PLANNING

PreAct: Prediction Enhances Agent’s Planning Ability (2025.coling-main)

Challenge:	This tutorial explores how planning has been learned and deployed in creative workflows . many human creative tasks involve extensive planning, and actions need to be taken .
Approach:	This tutorial explores how planning has been learned and deployed in creative workflows . authors discuss forward and backward learning approaches for planning in LLMs - and evaluation metrics tailored to latent plans .
Outcome:	This tutorial examines how planning has been learned and deployed in creative workflows . it discusses forward and backward learning approaches for planning in LLMs - evaluation metrics tailored to latent plans .

Copied to clipboard

Dayuan Fu, Jianzhao Huang, Siyuan Lu, Guanting Dong, Yejie Wang, Keqing He, Weiran Xu

WorkForceAgent-R1: Incentivizing Reasoning Capability in LLM-based Web Agents via Reinforcement Learning (2026.findings-eacl)

Challenge:	Existing methods to analyze Markov decision processes (MDPs) are based on chain-of-thought (COT) and historical thought, action, and observation.
Approach:	They propose a model that integrates prediction, reasoning, and action with other models to provide a wider range of reasoning and more efficient actions.
Outcome:	The proposed model outperforms the ReAct method in completing complex tasks and is more efficient when paired with other memory or selection strategy techniques.

Copied to clipboard

Yuchen Zhuang, Di Jin, Jiaao Chen, Wenqi Shi, Hanrui Wang, Chao Zhang

Assessing Logical Puzzle Solving in Large Language Models: Insights from a Minesweeper Case Study (2024.naacl-long)

Challenge:	Existing web agents relying on supervised fine-tuning struggle with generalization and robustness due to insufficient reasoning capabilities when handling the inherently dynamic nature of web interactions.
Approach:	They propose a large language model-empowered web agent that trains using a rule-based reinforcement learning framework to enhance single-step reasoning and planning for business-oriented web navigation tasks.
Outcome:	The proposed agent outperforms baseline LLM-based agents on the WorkArena benchmark by 10.26–16.59%.

Copied to clipboard

Yinghao Li, Haorui Wang, Chao Zhang

AIPOM: Agent-aware Interactive Planning for Multi-Agent Systems (2025.emnlp-demos)

Challenge:	Large Language Models (LLMs) have shown remarkable proficiency in language understanding and have been successfully applied to a variety of real-world tasks through task-specific fine-tuning or prompt engineering.
Approach:	They propose a task that challenges LLMs to identify the locations of mines based on numerical clues provided by adjacent cells.
Outcome:	The proposed task requires an understanding of each cell’s state, discerning spatial relationships between clues and mines, and strategizing actions based on logical deductions drawn from the arrangement of the cells.

Copied to clipboard

Hannah Kim, Kushan Mitra, Chen Shen, Dan Zhang, Estevam Hruschka

Challenge:	Large language models (LLMs) are being used for planning in orchestrated multi-agent systems . existing LLMs fall short of human expectations and lack effective mechanisms for users to inspect, understand, and control their behaviors.
Approach:	They propose a system supporting human-in-the-loop planning through conversational and graph-based interfaces.
Outcome:	AIPOM enables users to transparently inspect, refine, and collaboratively guide LLM-generated plans, significantly enhancing user control and trust in multi-agent workflows.

TeachMaster: Generative Teaching via Code (2026.acl-industry)

Copied to clipboard

Yuheng Wang, Runde Yang, Lin Wu, Jie Zhang, Jingru Fan, Tianle Zhou, Ruoyu Fu, Huatao Li, Ruijie Shi, Siheng Chen, Weinan E, Chen Qian

Learning the Effects of Physical Actions in a Multi-modal Environment (2023.findings-eacl)

Challenge:	Existing methods for creating video content are limited by high costs and slow update cycles.
Approach:	They propose a paradigm shifting educators from manual creators to high-level directors who focus on pedagogical intents while agents handle execution.
Outcome:	The proposed framework reduces production costs to 0.3% of traditional course videos and provides a robust solution for scalable education.

Copied to clipboard

Gautier Dagan, Frank Keller, Alex Lascarides

Text-Blueprint: An Interactive Platform for Plan-based Conditional Generation (2023.eacl-demo)

Challenge:	Large Language Models (LLMs) are trained on large corpora of disembodied texts.
Approach:	They propose a multi-modal task of predicting the outcomes of actions solely from realistic sensory inputs (images and text). They extend an LLM to model latent representations of objects to better predict action outcomes in an environment.
Outcome:	The proposed model can capture commonsense when augmented with visual information and generalize and learn commonsensical reasoning better.

Copied to clipboard

Fantine Huot, Joshua Maynez, Shashi Narayan, Reinald Kim Amplayo, Kuzman Ganchev, Annie Priyadarshini Louis, Anders Sandholm, Dipanjan Das, Mirella Lapata

Learning Planning-based Reasoning by Trajectories Collection and Process Reward Synthesizing (2024.emnlp-main)

Challenge:	Recent work shows that conditional generation models can be useful to control the text generation process, leading to irrelevant, repetitive, and hallucinated content.
Approach:	They propose a web browser-based demonstration for query-focused summarization that uses a sequence of question-answer pairs as a blueprint plan for guiding text generation.
Outcome:	The proposed model can be used to generate query-focused summarization text using question-answer pairs as a blueprint plan.

Copied to clipboard

Fangkai Jiao, Chengwei Qin, Zhengyuan Liu, Nancy Chen, Shafiq Joty

WebSTAR: Scalable Data Synthesis for Computer Use Agents with Step-Level Filtering (2026.acl-long)

Challenge:	Recent studies have raised concerns regarding the hallucination and flaws in their reasoning process.
Approach:	They propose a framework to learn planning-based reasoning through Direct Preference Optimization on collected trajectories, which are ranked according to synthesized process rewards.
Outcome:	The proposed model surpasses GPT-3.5-Turbo on logical reasoning benchmarks on a set of logically-based reasoning tasks.

Copied to clipboard

Yifei He, Pranit Chawla, Yaser Souri, Subhojit Som, Xia Song

DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environments (2025.emnlp-main)

Challenge:	Existing datasets rely on human demonstrations, limiting scalability.
Approach:	They propose a scalable data synthesis pipeline that transforms noisy rollouts into reliable supervision without human annotation.
Outcome:	The proposed pipeline transforms noisy rollouts into reliable supervision without human annotation.

Copied to clipboard

Yuxiang Zheng, Dayuan Fu, Xiangkun Hu, Xiaojie Cai, Lyumanshan Ye, Pengrui Lu, Pengfei Liu

Naver Labs Europe’s Systems for the Document-Level Generation and Translation Task at WNGT 2019 (D19-56)

Challenge:	Large Language Models (LLMs) with web search capabilities show significant potential for deep research.
Approach:	They introduce a framework for end-to-end training of LLM-based deep research agents . they implement a specialized multi-agent architecture where browsing agents extract relevant information from various webpage structures.
Outcome:	The proposed framework improves on open-domain research tasks by 28.9 points over prompt engineering and 7.2 points over RAG-based RL agents.

Copied to clipboard

Fahimeh Saleh, Alexandre Berard, Ioan Calapodescu, Laurent Besacier

ZeroSumEval: An Extensible Framework For Scaling LLM Evaluation with Inter-Model Competition (2025.acl-demo)

Challenge:	Recent advances in machine translation and natural language generation have created many challenges in this field especially when context is considered.
Approach:	They propose to leverage data from machine translation and natural language generation tasks to do transfer learning between MT, NLG and MT with source-side metadata.
Outcome:	The proposed approach outperforms the previous state-of-the-art on the Rotowire NLG task.

Copied to clipboard

Hisham Abdullah Alyahya, Haidar Khan, Yazeed Alnumay, M Saiful Bari, Bulent Yener

Challenge:	Recent studies highlight the effectiveness of game-based evaluations for Large Language Models.
Approach:	They propose a dynamic, competition-based evaluation framework for Large Language Models that leverages competitive games.
Outcome:	The framework leverages competitive games to evaluate models in large language models.

Learning Geolocations for Cold-Start and Hard-to-Resolve Addresses via Deep Metric Learning (2022.emnlp-industry)

Copied to clipboard

null Govind, Saurabh Sohoney

Representation Learning

Challenge:	Existing systems for learning geolocation fail to cater to a significant fraction of addresses which are new in the system and have inaccurate or missing building level information.
Approach:	They propose a framework to resolve addresses to a shallower granularity termed neighbourhood . they propose 'deep metric learning' model to encode geospatial semantics in address embeddings .
Outcome:	The proposed framework reduces delivery defects and delivery defects in India and the United Arab Emirates.

TinyScientist: An Interactive, Extensible, and Controllable Framework for Building Research Agents (2025.emnlp-demos)

Copied to clipboard

Haofei Yu, Keyang Xuan, Fenghai Li, Kunlun Zhu, Zijie Lei, Jiaxun Zhang, Ziheng Qi, Kyle Richardson, Jiaxuan You

MIMIR: A Customizable Agent Tuning Platform for Enhanced Scientific Applications (2024.emnlp-demo)

Challenge:	Existing research systems often design and use agentic workflows to perform research tasks such as ideation, scientific coding, review writing, and tree-based search.
Approach:	They propose an open-source codebase, an interactive web demonstration, and a PyPI Python package to make state-of-the-art auto-research pipelines broadly accessible to every researcher and developer.
Outcome:	The proposed framework adapts easily to new tools and supports iterative growth.

Copied to clipboard

Xiangru Tang, Chunyuan Deng, Hanminwang Hanminwang, Haoran Wang, Yilun Zhao, Wenqi Shi, Yi Fung, Wangchunshu Zhou, Jiannan Cao, Heng Ji, Arman Cohan, Mark Gerstein

OVM, Outcome-supervised Value Models for Planning in Mathematical Reasoning (2024.findings-naacl)

Challenge:	Large language models (LLMs) have evolved into interactive agents capable of planning, tool use, and task execution across various tasks.
Approach:	They propose a platform that leverages large language models to generate agent-tuning data for fine-tuneing smaller, specialized models.
Outcome:	MIMIR enables large models to simulate various roles and create interaction data, which can then be used to fine-tune open-source models like LLaMA2.

Copied to clipboard

Fei Yu, Anningzhe Gao, Benyou Wang

Target-Guided Open-Domain Conversation Planning (2022.coling-1)

Challenge:	Large language models (LLMs) struggle with maintaining accuracy throughout multiple reasoning steps, especially in mathematical reasoning where an error in earlier steps can propagate to subsequent ones and ultimately leading to an incorrect answer.
Approach:	They propose an Outcome-supervised Value Model (OVM) that employs outcome supervision for training a value model, which prioritizes steps that lead to accurate conclusions.
Outcome:	The proposed model performs better on two multi-step reasoning datasets, GSM8K and Game of 24.

Copied to clipboard

Yosuke Kishinami, Reina Akama, Shiki Sato, Ryoko Tokuhisa, Jun Suzuki, Kentaro Inui

Conditional Generation with a Question-Answering Blueprint (2023.tacl-1)

Challenge:	Existing studies on goal-oriented conversational tasks lack planning . prior studies on this topic have focused on generating proactive behavior in agents .
Approach:	They propose a task to evaluate whether neural conversational agents have goal-oriented conversation planning abilities.
Outcome:	The proposed task evaluates whether neural conversational agents have goal-oriented conversation planning abilities.

Copied to clipboard

Shashi Narayan, Joshua Maynez, Reinald Kim Amplayo, Kuzman Ganchev, Annie Louis, Fantine Huot, Anders Sandholm, Dipanjan Das, Mirella Lapata

Question Answering

Challenge:	Neural generation models often struggle to identify which content units are salient.
Approach:	They propose a new conceptualization of text plans as a sequence of question-answer pairs . they propose QA blueprints as QA proxy for content selection and planning .
Outcome:	The proposed model improves existing datasets with QA blueprints as proxy for content selection and planning.

PAIR: Planning and Iterative Refinement in Pre-trained Transformers for Long Text Generation (2020.emnlp-main)

Copied to clipboard

Xinyu Hua, Lu Wang

From Grounding to Manipulation: Case Studies of Foundation Model Integration in Embodied Robotic Systems (2025.findings-emnlp)

Challenge:	We present a content-controlled text generation framework for pre-trained Transformers . large pre-train models are the cornerstone of many state-of-the-art models in natural language understanding and generation tasks.
Approach:	They propose a content-controlled text generation framework that adds content planning to large pre-trained Transformers without modifying model architecture.
Outcome:	The proposed framework improves the quality of the outputs on three domains.

Copied to clipboard

Xiuchao Sui, Daiying Tian, Qi Sun, Ruirui Chen, Dongkyu Choi, Kenneth Kwok, Soujanya Poria

Dynamic Planning for LLM-based Graphical User Interface Automation (2024.findings-emnlp)

Challenge:	a new study examines the operational characteristics of different integration strategies for robotics . end-to-end vision-language-action models implicitly unify perception and planning .
Approach:	They propose end-to-end vision-language-action models that implicitly unify perception and planning . they also propose modular pipelines using either vision-linguistic models or MLLMs .
Outcome:	The proposed frameworks implicitly unify perception and planning, and modular pipelines using either vision-language models or multimodal large language models.

Copied to clipboard

Shaoqing Zhang, Zhuosheng Zhang, Kehai Chen, Xinbei Ma, Muyun Yang, Tiejun Zhao, Min Zhang

User Feedback Alignment for LLM-powered Exploration in Large-scale Recommendation Systems (2025.acl-industry)

Challenge:	Existing approaches to planning for GUI tasks are limited due to long historical dialogues.
Approach:	They propose a novel approach to dynamic planning based on environmental feedback and execution history to guide action prediction in GUI tasks.
Outcome:	The proposed approach surpasses the strong GPT-4V baseline by +12.7% in accuracy.

Copied to clipboard

Jianling Wang, Yifan Liu, Yinghao Sun, Xuejian Ma, Yueqi Wang, He Ma, Zhengyang Su, Minmin Chen, Mingyan Gao, Onkar Dalal, Ed H. Chi, Lichan Hong, Ningren Han, Haokai Lu

Data-Centric Perspectives on Agentic Retrieval-Augmented Generation: A Survey (2026.findings-acl)

Challenge:	Large Language Models (LLMs) can be used to broaden user experiences beyond established preferences and reinforce feedback loops.
Approach:	They propose a hierarchical approach that combines hierarchic planning with LLM inference-time scaling to improve recommendation relevancy without compromising novelty.
Outcome:	The proposed approach shows significant gains in both user satisfaction and exploration diversity.

Copied to clipboard

Jingwen Deng, Jihao Huang, Zhen Hao Wong, Hao Liang, Quanqing Xu, Bin Cui, Wentao Zhang

Information Retrieval

Challenge:	Large Language Models (LLMs) excel at natural language understanding and generation, yet rely on static pre-training data.
Approach:	They propose to augment Large Language Models with external retrieval to ground model outputs . traditional RAG is constrained by a fixed retrieve-then-generate routine . authors aim to guide creation of high-quality datasets for next generation of adaptive LLM agents .
Outcome:	The proposed model can decompose tasks, issue exploratory queries, and refine evidence through iterative retrieval.

Towards a Progression-Aware Autonomous Dialogue Agent (2022.naacl-main)

Copied to clipboard

Abraham Sanders, Tomek Strzalkowski, Mei Si, Albert Chang, Deepanshu Dey, Jonas Braasch, Dakuo Wang

PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR (2026.eacl-long)

Challenge:	Recent advances in large-scale language modeling and generation have enabled the creation of dialogue agents that exhibit human-like responses in a wide range of conversational scenarios.
Approach:	They propose a framework in which dialogue agents can evaluate the progression of a conversation toward or away from desired outcomes and use this signal to inform planning for subsequent responses.
Outcome:	The proposed framework evaluates the progression of a conversation toward or away from desired outcomes and uses this signal to inform planning for subsequent responses.

Copied to clipboard

James Burgess, Jan N. Hansen, Duo Peng, Yuhui Zhang, Alejandro Lozano, Min Woo Sun, Emma Lundberg, Serena Yeung-Levy

Information Retrieval

Challenge:	Recent methods supervise only the final answer accuracy using reinforcement learning with verifiable rewards (RLVR).
Approach:	They propose to train search agents to search and reason over scientific papers and a factoid QA dataset with 60k biomedical paper abstracts.
Outcome:	The proposed model outperforms non-RL retrieval baselines and is scalable and extendable to other scientific domains.

ItiNera: Integrating Spatial Optimization with Large Language Models for Open-domain Urban Itinerary Planning (2024.emnlp-industry)

Copied to clipboard

Yihong Tang, Zhaokai Wang, Ao Qu, Yihao Yan, Zhaofeng Wu, Dingyi Zhuang, Jushi Kai, Kebing Hou, Xiaotong Guo, Jinhua Zhao, Zhan Zhao, Wei Ma

Embodied Executable Policy Learning with Language-based Scene Summarization (2024.naacl-long)

Challenge:	Existing urban itinerary planning studies focus on traditional tourism, but they lack the precision and accuracy needed to create a personalized itinerary.
Approach:	They propose an open-domain urban itinerary planning system that integrates spatial optimization with large language models to provide customized urban itineraries based on user needs.
Outcome:	The proposed system can generate personalized urban itineraries based on user needs and scale with existing methods.

Copied to clipboard

Jielin Qiu, Mengdi Xu, William Han, Seungwhan Moon, Ding Zhao

Information Extraction & Text Mining

Challenge:	Existing Large Language models with text inputs lack the capability to evolve with non-expert interactions with environments.
Approach:	They propose a novel learning paradigm that generates robots’ executable actions in the form of text, derived solely from visual observations.
Outcome:	The proposed learning paradigm surpasses baselines and can adapt to the target tasks effectively.

Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding (2026.acl-long)

Copied to clipboard

Yuhang Zhou, Mingrui Zhang, Ke Li, Mingyi Wang, Qiao Liu, Qifei Wang, Jiayi Liu, Fei Liu, Serena Li, Weiwei LI, Mingze Gao, Abhishek Kumar, Xiangjun Fan, Zhuokai Zhao, Lizhu Zhang

Encouraging Good Processes Without the Need for Good Answers: Reinforcement Learning for LLM Agent Planning (2025.emnlp-industry)

Challenge:	Large language models (LLMs) have shown promise on understanding and reasoning over tables, but current approaches remain limited.
Approach:	They propose a multi-agent framework that decomposes table reasoning into three specialized roles: planning, coding, and answering.
Outcome:	The proposed framework decomposes table reasoning into three specialized roles: planning, coding, and answering.

Copied to clipboard

Zhiwei Li, Yong Hu, Wenqing Wang

Challenge:	Currently, the dominant end-to-end reinforcement learning paradigm for agents in Large Language Models (LLMs) employs multi-objective optimization that jointly trains both planning and answer summarization capabilities.
Approach:	They propose a framework that decouples the training process to enable a focused, single-objective optimization of the planning module.
Outcome:	The proposed framework achieves an 8%–12% improvement in planning performance compared to end-to-end baselines.

A Language-First Approach for Procedure Planning (2023.findings-acl)

Copied to clipboard

Jiateng Liu, Sha Li, Zhenhailong Wang, Manling Li, Heng Ji

Hybrid Semantics for Goal-Directed Natural Language Generation (2022.acl-long)

Challenge:	Developing intelligent agents requires the ability to produce plans on the fly based on visual observations.
Approach:	They propose a language-first procedure planning framework with a modularized design . they first align current and goal observations with corresponding steps and then use a pre-trained LM to predict intermediate steps.
Outcome:	The proposed framework matches state-of-the-art procedures on COIN and CrossTask benchmarks.

Copied to clipboard

Connor Baumler, Soumya Ray

AgentTuning: Enabling Generalized Agent Abilities for LLMs (2024.findings-acl)

Challenge:	Existing goal-directed natural language generation systems use first-order logic to represent semantics, but they are often slow due to the semantics of the partially realized text being checked.
Approach:	They propose to use logical semantics and distributional semantics to combine meaning representations to scale a goal-directed natural language generation system without losing expressiveness.
Outcome:	The proposed approach scales significantly better than the goal-directed generation system, but it is slower because the representations are not as precise as pure logical semantics.

Copied to clipboard

Aohan Zeng, Mingdao Liu, Rui Lu, Bowen Wang, Xiao Liu, Yuxiao Dong, Jie Tang

API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs (2023.emnlp-main)

Challenge:	Open large language models (LLMs) with great performance in various tasks are far inferior to commercial models such as ChatGPT and GPT-4 when acting as agents to tackle complex tasks in the real world.
Approach:	They propose a method to enhance the agent capabilities of LLMs while maintaining their general abilities.
Outcome:	The AgentLM-70B is comparable to GPT-3.5-turbo on unseen agent tasks, demonstrating generalized agent capabilities.

Copied to clipboard

Minghao Li, Yingxiu Zhao, Bowen Yu, Feifan Song, Hangyu Li, Haiyang Yu, Zhoujun Li, Fei Huang, Yongbin Li

LUMINA: Long-horizon Understanding for Multi-turn Interactive Agents (2026.findings-acl)

Challenge:	Recent research has demonstrated that Large Language Models (LLMs) can enhance their capabilities by utilizing external tools.
Approach:	They propose a runnable evaluation system consisting of 73 API tools and an annotation system for 314 tool-use dialogues with 753 API calls.
Outcome:	The proposed benchmark assesses the effectiveness of existing LLMs by analyzing 314 tool-use dialogues with 753 API calls.

Copied to clipboard

Amin Rakhsha, Thomas Hehn, Pietro Mazzaglia, Fabio Valerio Massoli, Arash Behboodi, Tribhuvanesh Orekondy

Deep Dyna-Q: Integrating Planning for Task-Completion Dialogue Policy Learning (P18-1)

Challenge:	Large language models struggle on multi-turn, long-horizon agentic problems that require skills such as planning, state tracking, and long context processing.
Approach:	They propose an oracle counterfactual framework for multi-turn problems that asks: how would an agent perform if it could leverage an or acle to execute a specific skill?
Outcome:	The proposed framework allows for precise oracle interventions without confounding effects present in real-world benchmarks.

Copied to clipboard

Baolin Peng, Xiujun Li, Jianfeng Gao, Jingjing Liu, Kam-Fai Wong

KnowAgent: Knowledge-Augmented Planning for LLM-Based Agents (2025.findings-naacl)

Challenge:	Training a task-completion dialogue agent via reinforcement learning (RL) is costly because it requires many interactions with real users.
Approach:	They propose a framework that integrates planning for task-completion dialogue policy learning into a dialogue agent using a world model to mimic real user response and generate simulated experience.
Outcome:	The proposed framework integrates planning for task-completion dialogue policy learning with real user interaction and simulated user behavior.

Copied to clipboard

Yuqi Zhu, Shuofei Qiao, Yixin Ou, Shumin Deng, Shiwei Lyu, Yue Shen, Lei Liang, Jinjie Gu, Huajun Chen, Ningyu Zhang

Autonomous Workflow for Multimodal Fine-Grained Training Assistants Towards Mixed Reality (2024.findings-acl)

Challenge:	Large Language Models (LLMs) fail to effectively guide the planning trajectories during task solving and result in planning hallucinations.
Approach:	They propose a novel approach to enhance the planning capabilities of large language models by incorporating explicit action knowledge.
Outcome:	The proposed approach can achieve comparable or superior performance to existing baselines on HotpotQA and ALFWorld.

Copied to clipboard

Jiahuan Pei, Irene Viola, Haochen Huang, Junxiao Wang, Moonisa Ahsan, Fanghua Ye, Jiang Yiming, Yao Sai, Di Wang, Zhumin Chen, Pengjie Ren, Pablo Cesar

Planning, Creation, Usage: Benchmarking LLMs for Comprehensive Tool Utilization in Real-World Complex Scenarios (2024.findings-acl)

Challenge:	a fine-grained, comprehensive understanding of multimodal environments remains under-explored.
Approach:	They propose an automated workflow for integrating AI agents into extended reality (XR) they propose a cerebral language agent that integrates LLM with memory, planning, and interaction with XR tools and a vision-language agent .
Outcome:	The proposed workflow integrates AI agents seamlessly into extended reality (XR) applications for fine-grained training.

Copied to clipboard

Shijue Huang, Wanjun Zhong, Jianqiao Lu, Qi Zhu, Jiahui Gao, Weiwen Liu, Yutai Hou, Xingshan Zeng, Yasheng Wang, Lifeng Shang, Xin Jiang, Ruifeng Xu, Qun Liu

ADaPT: As-Needed Decomposition and Planning with Language Models (2024.findings-naacl)

Challenge:	Existing benchmarks focus on simple synthesized queries that do not reflect real-world complexity, thereby offering limited perspectives in evaluating tool utilization.
Approach:	They propose a benchmark to evaluate LLMs’ ability in tool utilization within real-world scenarios.
Outcome:	The proposed benchmark improves LLMs’ ability in tool utilization within real-world scenarios and eliminates the restriction of pre-defined toolset.

Copied to clipboard

Archiki Prasad, Alexander Koller, Mareike Hartmann, Peter Clark, Ashish Sabharwal, Mohit Bansal, Tushar Khot

RecMind: Large Language Model Powered Agent For Recommendation (2024.findings-naacl)

Challenge:	Large Language Models (LLMs) are increasingly being used for interactive decision-making tasks requiring planning and adapting to the environment.
Approach:	They propose an approach that explicitly plans and decomposes complex sub-tasks when the LLM is unable to execute them.
Outcome:	The proposed approach significantly outperforms established strong baselines, achieving success rates up to 28.3% higher in ALFWorld, 27% in WebShop, and 33% in TextCraft.

Copied to clipboard

Yancheng Wang, Ziyan Jiang, Zheng Chen, Fan Yang, Yingxue Zhou, Eunah Cho, Xing Fan, Yanbin Lu, Xiaojiang Huang, Yingzhen Yang

On Evaluating the Integration of Reasoning and Action in LLM Agents with Database Question Answering (2024.findings-naacl)

Challenge:	Existing recommendations systems are limited in generalizing to new tasks due to model scale and data size constraints.
Approach:	They propose an LLM-powered autonomous recommender agent, RecMind, which is capable of leveraging external knowledge to provide zero-shot personalized recommendations.
Outcome:	The proposed model outperforms existing zero/few-shot LLM-based recommendation baseline methods in various tasks and achieves comparable performance to a fully trained recommendation model P5.

Copied to clipboard

Linyong Nan, Ellen Zhang, Weijin Zou, Yilun Zhao, Wenfei Zhou, Arman Cohan

CogGen: A Cognitively Inspired Recursive Framework for Deep Research Report Generation (2026.findings-acl)

Challenge:	a new study evaluates how Large Language Models interact with a SQL interpreter . the model is limited in context and is stochastic, making it less suited for tasks requiring high precision and extensive computations.
Approach:	They propose and evaluate two interaction strategies to evaluate how LLMs interact with a SQL interpreter.
Outcome:	The proposed framework improves the accuracy and reliability of the evaluations.

Copied to clipboard

Kuo Tian, Pengfei Sun, Zhen Wu, Junran Ding, Xinyu Dai

MacGyver: Are Large Language Models Creative Problem Solvers? (2024.naacl-long)

Challenge:	Existing approaches to deep research report generation rely on rigid predefined linear workflows, which cause error accumulation and limit in-depth multimodal fusion and report quality.
Approach:	They propose a Cognitively inspired recursive framework for deep research report Generation that simulates cognitive writing and abstract visual representation (AVR) they also propose CLEF, a cognitive load evaluation framework, and a benchmark from our world in data.
Outcome:	The proposed framework achieves state-of-the-art among open-source systems, surpassing Gemini Deep Research.

Copied to clipboard

Yufei Tian, Abhilasha Ravichander, Lianhui Qin, Ronan Le Bras, Raja Marjieh, Nanyun Peng, Yejin Choi, Thomas Griffiths, Faeze Brahman

Hephaestus: Improving Fundamental Agent Capabilities of Large Language Models through Continual Pre-Training (2025.naacl-long)

Challenge:	a new study examines the creative problem-solving capabilities of modern LLMs . it provides insight into the constrained problem- solving capabilities of both humans and AI .
Approach:	They use an automatically generated dataset to compare and contrast LLMs and humans to find out their creative problem-solving abilities.
Outcome:	The proposed dataset compares LLMs and humans in a constrained setting . it shows that humans excel in tasks they are familiar with but struggle with domain-specific knowledge .

Copied to clipboard

Yuchen Zhuang, Jingfeng Yang, Haoming Jiang, Xin Liu, Kewei Cheng, Sanket Lokegaonkar, Yifan Gao, Qing Ping, Tianyi Liu, Binxuan Huang, Zheng Li, Zhengyang Wang, Pei Chen, Ruijie Wang, Rongzhi Zhang, Nasser Zalmout, Priyanka Nigam, Bing Yin, Chao Zhang

Debate-to-Write: A Persona-Driven Multi-Agent Framework for Diverse Argument Generation (2025.coling-main)

Challenge:	Existing LLMs often rely on complex prompting or extensive fine-tuning to introduce new capabilities while preserving strong generalizability.
Approach:	They propose a large-scale pre-training corpus to enhance LLM agents' capabilities . they use 103B agent-specific data encompassing 76,537 APIs .
Outcome:	The proposed training corpus outperforms open-source LLMs and commercial LLM agents on three agent benchmarks.

Copied to clipboard

Zhe Hu, Hou Pong Chan, Jing Li, Yu Yin

Challenge:	Writing arguments requires integrating high-level beliefs from various perspectives . current language models generate outputs autoregressively, resulting in limited diversity and coherence .
Approach:	They propose a persona-based multi-agent framework for argument writing that integrates beliefs from different perspectives into a coherent narrative.
Outcome:	The proposed framework generates more diverse arguments by both automatic and human evaluations.

BAR: A Backward Reasoning based Agent for Complex Minecraft Tasks (2025.findings-acl)

Copied to clipboard

Weihong Du, Wenrui Liao, Binyu Yan, Hongru Liang, Anthony G Cohn, Wenqiang Lei

Long and Diverse Text Generation with Planning-based Hierarchical Variational Model (D19-1)

Challenge:	Existing studies focus on forward reasoning based planning, but this paradigm doesn't work well for complex tasks.
Approach:	They propose to decompose a task into easily executed steps by planning and use a backward reasoning based agent to make the planning starting from the terminal state.
Outcome:	The proposed model outperforms existing methods and the proposed modules in a virtual environment that simulates complex tasks based on real-world scenarios.

Copied to clipboard

Zhihong Shao, Minlie Huang, Jiangtao Wen, Wenfei Xu, Xiaoyan Zhu

Navigating Large-Scale Document Collections: MuDABench for Multi-Document Analytical QA (2026.findings-acl)

Challenge:	Existing methods for data-to-text generation are insufficient to produce long and diverse texts.
Approach:	They propose a planning-based hierarchical variational model that plans a sequence of groups and then realizes each sentence conditioned on the planning result and the previously generated context.
Outcome:	The proposed model outperforms state-of-the-art models in long and diverse text generation.

Copied to clipboard

Zhanli Li, Yixuan Cao, Lvzhou Luo, Ping Luo

Information Extraction & Text Mining

Challenge:	Existing multi-document QA benchmarks require information from only a few documents with limited cross-document reasoning.
Approach:	They propose a benchmark for multi-document analytical QA that extracts and synthesizes information across multiple documents to perform quantitative analysis.
Outcome:	The proposed approach improves both process and outcome metrics but still has bottlenecks compared to human experts.

Verification-Aware Planning for Multi-Agent Systems (2026.eacl-long)

Copied to clipboard

Tianyang Xu, Dan Zhang, Kushan Mitra, Estevam Hruschka

Challenge:	Large language model (LLM) agents are increasingly deployed to tackle complex tasks . multi-agent collaboration introduces new challenges in planning, coordination, and verification .
Approach:	They propose a framework for multi-agent collaboration with verification-aware planning . the framework decomposes tasks, models subtask dependencies, and encodes planner-defined passing criteria as subtask verification functions (VFs)
Outcome:	The proposed framework outperforms baselines on diverse datasets while improving system robustness and interpretability.

Attribution-Based Analysis and Optimization of Modular Agentic Workflows (2026.findings-acl)

Copied to clipboard

Yingxuan Yang, Bo Huang, Siyuan Qi, Chao Feng, Haoyi Hu, Yuxuan Zhu, Jinbo Hu, Haoran Zhao, Ziyi He, Xiao Liu, ZongYu Wang, Muning Wen, Lin Qiu, Xuezhi Cao, Xunliang Cai, Yong Yu, Weinan Zhang

Challenge:	Large Language Models (LLMs) have driven the rise of agentic workflows . yet, how can we attribute performance gains to individual upgrades and their interactions?
Approach:	They propose a game-theoretic framework that models component upgrades as players and evaluates component coalitions to compute Shapley values.
Outcome:	The proposed framework provides interaction-aware attribution and recommendation for model allocation under a fixed workflow structure.

SolidCoder: Bridging the Mental-Reality Gap in LLM Code Generation through Concrete Execution (2026.findings-acl)

Copied to clipboard

Woojin Lee, Jin-Xia Huang

Programming Languages in NLP

Challenge:	State-of-the-art code generation frameworks rely on mental simulations to validate buggy code.
Approach:	They propose a mental-reality gap between mental simulation and actual execution . they propose sandboxed execution with a simple principle: don't imagine—execute .
Outcome:	The proposed framework achieves state-of-the-art pass@1 performance on humanEval, CodeContests and APPS.

Visual Storytelling with Question-Answer Plans (2023.findings-emnlp)

Copied to clipboard

Danyang Liu, Mirella Lapata, Frank Keller

Unifying Inference-Time Planning Language Generation (2026.findings-acl)

Challenge:	Existing models focus on enhancing the representation of image sequences, but the stories are repetitive, illogical, and lacking in detail.
Approach:	They propose a framework which integrates visual representations with pretrained language models and planning.
Outcome:	The proposed framework combines visual representations with pretrained language models and planning.

Copied to clipboard

Prabhu Prakash Kagitha, Bo Sun, Ishan Desai, Andrew Zhu, Cassie Huang, Manling Li, Ziyang Li, Li Zhang

Discriminative Deep Dyna-Q: Robust Planning for Dialogue Policy Learning (D18-1)

Challenge:	Large language models (LLMs) are used to generate a formal representation of a plan in a planning language.
Approach:	They propose a unifying organizational framework based on intermediate representations to unify the inference-time LLM-as-formalizer methodology for classical planning.
Outcome:	The proposed framework subsumes most existing work and proposes new ones that involve syntactically similar but high-resource intermediate languages.

Copied to clipboard

Shang-Yu Su, Xiujun Li, Jianfeng Gao, Jingjing Liu, Yun-Nung Chen

MAgIC: Investigation of Large Language Model Powered Multi-Agent in Cognition, Adaptability, Rationality and Collaboration (2024.emnlp-main)

Challenge:	Existing approaches to improve the effectiveness and robustness of Deep Dyna-Q (DDQ) are based on a discriminator to control the quality of simulated experiences and to improve learning.
Approach:	They propose to use an RNN-based discriminator to control the quality of simulated experience to improve the effectiveness and robustness of Deep Dyna-Q.
Outcome:	The proposed framework outperforms DDQ by controlling the quality of simulated experience used for planning.

Copied to clipboard

Lin Xu, Zhiyuan Hu, Daquan Zhou, Hongyu Ren, Zhen Dong, Kurt Keutzer, See-Kiong Ng, Jiashi Feng

Unlocking the Future: Exploring Look-Ahead Planning Mechanistic Interpretability in Large Language Models (2024.emnlp-main)

Challenge:	Large language models (LLMs) have advanced natural language processing, demonstrating exceptional reasoning, tool usage, and memory capabilities.
Approach:	They propose a competition-based benchmark framework specifically designed to assess LLMs within multi-agent environments.
Outcome:	The proposed framework enhances the LLMs’ abilities in navigating complex social and cognitive dimensions by over threefold between the strongest and weakest LLM models.

Copied to clipboard

Tianyi Men, Pengfei Cao, Zhuoran Jin, Yubo Chen, Kang Liu, Jun Zhao

FACTS: Table Summarization via Offline Template Generation with Agentic Workflows (2026.findings-acl)

Challenge:	Recent studies have shown that large language models may possess preliminary planning capabilities.
Approach:	They examine the look-ahead planning mechanism in large language models from the perspectives of information flow and internal representations.
Outcome:	The proposed model can decode the decision from the output of MHSA in the middle layers at the last token.

Copied to clipboard

Ye Yuan, Mohammad Amin Shabani, Siqi Liu

Dr3: Ask Large Language Models Not to Give Off-Topic Answers in Open Domain Multi-Hop Question Answering (2024.lrec-main)

Challenge:	Existing methods for query-focused table summarization struggle with complex reasoning and token-limit issues.
Approach:	They propose a Fast, Accurate, and Privacy-Compliant table summarization approach via Offline Template Generation.
Outcome:	The proposed method outperforms baseline methods on widely-used benchmarks.

Copied to clipboard

Yuan Gao, Yiheng Zhu, Yuanbin Cao, Yinzhi Zhou, Zhen Wu, Yujie Chen, Shenglan Wu, Haoyuan Hu, Xinyu Dai

Question Answering

Challenge:	Open Domain Multi-Hop Question Answering (ODMHQA) is one of the most challenging tasks in Natural Language Processing (NLP)
Approach:	They propose a mechanism that leverages the intrinsic capabilities of Large Language Models to judge whether the generated answers are off-topic.
Outcome:	The proposed method reduces the occurrence of off-topic answers by nearly 13%, improving the performance in Exact Match (EM) by nearly 3% compared to the baseline method without the Dr3 mechanism.

On the Multi-turn Instruction Following for Conversational Web Agents (2024.acl-long)

Copied to clipboard

Yang Deng, Xuan Zhang, Wenxuan Zhang, Yifei Yuan, See-Kiong Ng, Tat-Seng Chua

Debt Collection Negotiations with Large Language Models: An Evaluation System and Optimizing Decision Making with Multi-Agent (2025.findings-acl)

Challenge:	Large Language Models (LLMs) have demonstrated remarkable abilities in planning and executing multi-step interactions within web-based environments.
Approach:	They propose a framework for conversational web navigation that uses multi-turn interactions with both the user and the environment.
Outcome:	The proposed framework is based on a multi-turn Mind2Web (MT-Mind2Web) it is designed to perform multi-step interactions with web-based environments .

Copied to clipboard

Xiaofeng Wang, Zhixin Zhang, Jin Guang Zheng, Yiming Ai, Rui Wang

Making Large Language Models into World Models with Precondition and Effect Knowledge (2025.coling-main)

Challenge:	Debt collection negotiations (DCN) are vital for managing non-performing loans (NPLs) prior systems lacking dynamic negotiation and real-time decision-making capabilities.
Approach:	They propose a framework for debt negotiation that incorporates planning and judging modules to improve decision rationality.
Outcome:	The proposed framework improves decision rationality and integrates planning and judging modules to improve decision rationalness.

Copied to clipboard

Kaige Xie, Ian Yang, John Gunerli, Mark Riedl

Learning from the Irrecoverable: Error-Localized Policy Optimization for Tool-Integrated LLM Reasoning (2026.acl-long)

Challenge:	Large Language Models (LLMs) are not inherently designed to model real-world dynamics, but can be induced to perform two critical world model functions: determining the applicability of an action based on a given world state and predicting the resulting world state upon action execution.
Approach:	They propose to use Large Language Models to model world states and preconditions . they validate that precondition and effect knowledge generated by LLMs aligns with human understanding of world dynamics .
Outcome:	The proposed model can predict valid actions and state transitions, thereby replicating existing models.

Copied to clipboard

Qiao Liang, Yuke Zhu, Chao Ge, Lei Yang, Ying Shen, Bo Zheng, Sheng Guo

A Reward-Guided Dual-Phase Framework for Adaptive Inference-Time Reasoning (2026.findings-acl)

Challenge:	Tool-integrated reasoning (TIR) enables LLM agents to solve tasks through planning, tool use, and iterative revision, but outcome-only reinforcement learning suffers from sparse, delayed rewards and weak step-level credit assignment.
Approach:	They propose a tool-integrated reasoning approach that localizes the first irrecoverable step and leverages it for fine-grained credit assignment.
Outcome:	The proposed algorithm outperforms strong Agentic RL benchmarks in math, science QA, and code execution with additional gains in Pass@K and Major@K scaling, rollout ranking quality, and tool-call efficiency.

Copied to clipboard

Yingqian Cui, Zhenwei Dai, Pengfei He, Bing He, Hui Liu, Zhan Shi, Xianfeng Tang, Jingying Zeng, Suhang Wang, Yue Xing, Jiliang Tang, Benoit Dumoulin

T-Eval: Evaluating the Tool Utilization Capability of Large Language Models Step by Step (2024.acl-long)

Challenge:	Large Language Models (LLMs) have made strong progress in reasoning.
Approach:	They propose a dual-phase test-time scaling framework that separates planning and execution and performs search over each phase independently.
Outcome:	Experiments on math reasoning and code generation benchmarks show that the proposed approach improves accuracy while reducing redundant computation.

Copied to clipboard

Zehui Chen, Weihua Du, Wenwei Zhang, Kuikun Liu, Jiangning Liu, Miao Zheng, Jingming Zhuo, Songyang Zhang, Dahua Lin, Kai Chen, Feng Zhao

QPaug: Question and Passage Augmentation for Open-Domain Question Answering of LLMs (2024.findings-emnlp)

Challenge:	Existing studies evaluate the tool utilization ability of large language models based on the final output or only consider the single-step tool calling.
Approach:	They propose a new approach to evaluate the tool utilization capability of large language models (LLMs) they decompose the tool usage into multiple sub-processes, including instruction following, planning, reasoning, retrieval, understanding, and review.
Outcome:	The proposed model exhibits consistency with the outcome-oriented evaluation and provides a more fine-grained analysis of the capabilities of LLMs.

Copied to clipboard

Minsang Kim, Cheoneum Park, Seung Baek

Information Retrieval

Challenge:	Existing approaches to augmented generation of retrieved passages rely on the quality of a question's retrieved information.
Approach:	They propose a simple yet efficient method called question and passage augmentation via LLMs for open-domain QA.
Outcome:	The proposed method outperforms the state-of-the-art and achieves significant performance gain over existing methods.

Efficient Test-Time Scaling of Multi-Step Reasoning by Probing Internal States of Large Language Models (2026.acl-long)

Copied to clipboard

Jingwei Ni, Ekaterina Fadeeva, Tianyi Wu, Mubashara Akhtar, Jiaheng Zhang, Elliott Ash, Markus Leippold, Timothy Baldwin, See-Kiong Ng, Artem Shelmanov, Mrinmaya Sachan

TrustAgent: Towards Safe and Trustworthy LLM-based Agents (2024.findings-emnlp)

Challenge:	Existing verification approaches, such as Process Reward Models, are computationally expensive and limited to specific domains.
Approach:	They propose a transformer-based probe that uses internal states of frozen LLMs to estimate credibility of reasoning steps during generation.
Outcome:	The proposed probes match or exceed PRMs that are up to 810 larger.

Copied to clipboard

Wenyue Hua, Xianjun Yang, Mingyu Jin, Zelong Li, Wei Cheng, Ruixiang Tang, Yongfeng Zhang

Automated Chess Commentator Powered by Neural Chess Engine (P19-1)

Challenge:	Existing LLMs are primarily used for simple text-related tasks, but LLM-based agents can undertake more complex tasks that require planning and interaction with the physical world and humans.
Approach:	They propose an Agent-Constitution-based agent framework with a particular focus on improving the LLM-based agents' safety.
Outcome:	The proposed framework can enhance an LLM agent’s safety across multiple domains by identifying and mitigating potential dangers during the planning process.

Copied to clipboard

Hongyu Zang, Zhiwei Yu, Xiaojun Wan

Learning to Plan and Generate Text with Citations (2024.acl-long)

Challenge:	Existing approaches to generate chess commentary are limited in template variety and are not precise enough.
Approach:	They propose a neural chess engine into text generation models to help with encoding boards, predicting moves, and analyzing situations.
Outcome:	The proposed model can be trained to generate chess commentary texts in 5 categories . the results are both automatic and human evaluations of the model .

Copied to clipboard

Constanza Fierro, Reinald Kim Amplayo, Fantine Huot, Nicola De Cao, Joshua Maynez, Shashi Narayan, Mirella Lapata

A Survey on LLM-powered Agents for Recommender Systems (2025.findings-emnlp)

Challenge:	Large language models (LLMs) are increasingly useful in information-seeking scenarios, ranging from answering simple questions to generating responses to search-like queries.
Approach:	They propose to use plan-based models to improve faithfulness, grounding, and controllability of generated content and its organization.
Outcome:	The proposed models improve faithfulness, grounding, and controllability of generated content and its organization.

Copied to clipboard

Qiyao Peng, Hongtao Liu, Hua Huang, Jian Yang, Qing Yang, Minglai Shao

Language Model as Planner and Formalizer under Constraints (2026.acl-long)

Challenge:	Large Language Models have demonstrated remarkable capabilities in natural language understanding, reasoning, and generation.
Approach:	They present a comprehensive synthesis of large language models and their applications . they dissect a four-module agent architecture and review representative designs .
Outcome:	The proposed models address fundamental challenges in traditional recommender systems . they include limited comprehension of complex user intents, insufficient interaction capabilities .

Copied to clipboard

Cassie Huang, Stuti Mohan, Ziyi Yang, Stefanie Tellex, Li Zhang

Ask-before-Plan: Proactive Language Agents for Real-World Planning (2024.findings-emnlp)

Challenge:	Large language models (LLMs) have been widely used in planning but lack interpretability and control.
Approach:	They propose to augment widely used planning benchmarks with manually annotated, fine-grained, and rich natural language constraints spanning four formally defined categories.
Outcome:	The proposed model outperforms existing models in 4 state-of-the-art reasoning LLMs, 4 formal languages, and 4 datasets.

Copied to clipboard

Xuan Zhang, Yang Deng, Zifeng Ren, See-Kiong Ng, Tat-Seng Chua

Challenge:	despite the advancements of large language models, the potential of LLM-powered agents to comprehend ambiguous user instructions is still under exploration.
Approach:	They propose a task that requires agents to predict clarification needs based on conversation and agentenvironment interaction and generate a plan to fulfill the user's demands.
Outcome:	The proposed framework is based on a new ask-before-plan benchmark dataset.

Large Language Models as Source Planner for Personalized Knowledge-grounded Dialogues (2023.findings-emnlp)

Copied to clipboard

Hongru Wang, Minda Hu, Yang Deng, Rui Wang, Fei Mi, Weichao Wang, Yasheng Wang, Wai-Chung Kwan, Irwin King, Kam-Fai Wong

A Review of Prominent Paradigms for LLM-Based Agents: Tool Use, Planning (Including RAG), and Feedback Learning (2025.coling-main)

Challenge:	Existing knowledge-grounded dialogue systems focus on a single knowledge source or ignore the dependency between multiple knowledge sources.
Approach:	They propose a framework that integrates multiple knowledge sources and dependencies between them.
Outcome:	The proposed framework can produce persona-consistent and knowledge-enhanced responses on a knowledge-grounded dialogue dataset.

Copied to clipboard

Xinzhe Li

Coarse-to-Fine Grounded Memory for LLM Agent Planning (2025.emnlp-main)

Challenge:	Large Language Models have been used for planning, tool use, and feedback learning . inconsistent taxonomy and complexity of workflows create challenges .
Approach:	They propose a unified taxonomy to review and discuss the three paradigms . they define environments/tasks, common LLM-profiled roles and universally applicable workflows based on prior work .
Outcome:	The proposed taxonomy compares LMPR implementations and workflow usage across paradigms . large language models have human-like reasoning capabilities, the authors say .

Copied to clipboard

Wei Yang, Jinwei Xiao, Hongming Zhang, Qingyang Zhang, Yanna Wang, Bo Xu

Challenge:	Existing methods to enhance LLM with offline experiences or online trajectory analysis focus on single-granularity memory derived from dynamic environmental interactions.
Approach:	They propose a framework that grounds coarse-to-fine memories with LLM to enable flexible adaptation to diverse scenarios.
Outcome:	Extensive experiments on AlfWorld, Webshop and ScienceWorld show that the proposed framework outperforms baselines and comprehensively optimizes memory-enhanced LLM Agent system.

LAM SIMULATOR: Advancing Data Generation for Large Action Model Training via Online Exploration and Trajectory Feedback (2025.findings-acl)

Copied to clipboard

Thai Quoc Hoang, Kung-Hsiang Huang, Shirley Kokane, Jianguo Zhang, Zuxin Liu, Ming Zhu, Jake Grigsby, Tian Lan, Michael S Ryoo, Chien-Sheng Wu, Shelby Heinecke, Huan Wang, Silvio Savarese, Caiming Xiong, Juan Carlos Niebles

Challenge:	Large Action Models (LAMs) face challenges due to the need for high-quality training data, especially for multi-steps tasks that involve planning, executing tool calls, and responding to feedback.
Approach:	They propose a framework for online exploration of agentic tasks with high-quality feedback . they use a dynamic task query generator and an extensive collection of tools to create a high-level feedback environment for LLM Agents.
Outcome:	The proposed framework achieves 49.3% performance improvement over baselines on toolbench and CRMArena.

Planning with Multi-Constraints via Collaborative Language Agents (2025.coling-main)

Copied to clipboard

Cong Zhang, Xin Deik Goh, Dexun Li, Hao Zhang, Yong Liu

Browsing Like Human: A Multimodal Web Agent with Experiential Fast-and-Slow Thinking (2025.acl-long)

Challenge:	Recent advances in neural language models have sparked a new surge of intelligent agent research.
Approach:	They propose a method for collaborative LLM-based multi-agent systems that simplifies complex task planning with constraints by decomposing it into a hierarchy of subordinate tasks.
Outcome:	The proposed method achieves an average success rate of 42.68% on two constraint-intensive benchmarks, TravelPlanner and API-Bank.

Copied to clipboard

Haohao Luo, Jiayi Kuang, Wei Liu, Ying Shen, Jian Luan, Yang Deng

MMAPG: A Training-Free Framework for Multimodal Multi-hop Question Answering via Adaptive Planning Graphs (2025.emnlp-main)

Challenge:	Existing web agents lack visual perception, planning, and memory abilities, but their reasoning process is deviate from human cognition.
Approach:	They propose a multimodal web agent framework that emulates human planning process to decompose complex user instructions.
Outcome:	The proposed framework emulates human planning process to decompose complex user instructions.

Copied to clipboard

Yiheng Hu, Xiaoyang Wang, Qing Liu, Xiwei Xu, Qian Fu, Wenjie Zhang, Liming Zhu

LLM Agents in Law: Taxonomy, Applications, and Challenges (2026.acl-long)

Challenge:	Existing multimodal question answering models rely on sequential retrieval and reasoning, but this single-path paradigm makes them vulnerable to errors due to misleading intermediate steps.
Approach:	They propose a multimodal multi-hop question answering framework guided by an Adaptive Planning Graph . they propose modality-specific strategies that dynamically adapt to distinct data types .
Outcome:	The proposed framework outperforms existing models that rely on training.

Copied to clipboard

Shuang Liu, Ruijia Zhang, Ruoyun Ma, Yujia Deng, Lanyi Zhu, Jiayu Li, Zelong Li, Zhibin Shen, Mengnan Du

Challenge:	Large language models (LLMs) have improved the legal domain, but deployment of standalone models faces significant limitations regarding hallucination, outdated information, and verifiability.
Approach:	They present a survey of LLM agents for legal tasks and analyze their architectures . they analyze the transition from standard legal LLMs to legal agents .
Outcome:	The proposed architectures bridge the gap between technical capabilities and domain-specific needs.

How Much Do Large Language Models Know about Human Motion? A Case Study in 3D Avatar Control (2025.findings-emnlp)

Copied to clipboard

Kunhang Li, Jason Naradowsky, Yansong Feng, Yusuke Miyao

BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents (2026.findings-acl)

Challenge:	a new study explores the human motion knowledge of Large Language Models (LLMs) using 3D avatar control.
Approach:	They use 20 representative motion instructions to interpolate LLMs into avatar animations . they find they are strong at interpreting high-level body movements but struggle with precise body part positioning .
Outcome:	The proposed model is strong at interpreting high-level body movements but struggles with precise body part positioning.

Evolving Agents (2026.acl-long)

Copied to clipboard

Leonardo Ranaldi

Challenge:	Current models are static entities incapable of compressing complexity of real world into generalisable concepts . authors: lack of endogenous mechanism for representation updating renders models vulnerable to domain mismatch and catastrophic forgetting .
Approach:	a meta-control system distils on-the-fly abstract representations of states, actions, goals . authors propose a paradigm for autonomous learning driven by pseudo-symbolic abstraction .
Outcome:	a meta-control system distils on-the-fly abstract representations of states, actions, goals . a novel approach resolves the domain mismatch problem and lays the groundwork for truly autonomous AI models .

Copied to clipboard

Yunhao Feng, Yige Li, Yutao Wu, Yingshui Tan, Yanming Guo, Yifan Ding, Kun Zhai, Xingjun Ma, Yu-Gang Jiang

mPresenter: An Agentic Framework for Generating Multilingual Presentation Videos from Scientific Papers (2026.findings-acl)

Challenge:	Large language model (LLM) agents execute tasks through multi-step workflows that combine planning, memory, and tool use.
Approach:	They propose a modular framework that provides a unified view of backdoor threats in LLM agents.
Outcome:	The proposed framework provides a unified, agent-centric view of backdoor threats in LLM agents.

Copied to clipboard

Wenhan Han, Xiao Xiao, Mykola Pechenizkiy, Meng Fang

Visual Data in NLP

Challenge:	Existing Paper2Video systems are monolingual and often rely on single-pass pipelines.
Approach:	They propose a multilingual agentic Paper2Video system that decomposes the task into planning, audience-oriented critique, layout-aware slide generation, and multilingual figure interpretation.
Outcome:	The proposed system improves question-answering accuracy relative to previous systems while maintaining affordable cost and latency.

BLADE: Benchmarking Language Model Agents for Data-Driven Science (2024.findings-emnlp)

Copied to clipboard

Ken Gu, Ruoxi Shang, Ruien Jiang, Keying Kuang, Richard-John Lin, Donghe Lyu, Yue Mao, Youran Pan, Teng Wu, Jiaqian Yu, Yikun Zhang, Tianmai Zhang, Lanyi Zhu, Mike Merrill, Jeffrey Heer, Tim Althoff

Model-Based Imaginative Planning for Embodied Agents (2026.acl-long)

Challenge:	Language model-based agents can be used to conduct and support data-driven science, but evaluating them on open-ended tasks is challenging due to multiple valid approaches, partially correct steps, and different ways to express the same decisions.
Approach:	They propose a benchmark to automatically evaluate agents’ multifaceted approaches to open-ended research questions.
Outcome:	BLADE evaluates agents’ multifaceted approaches to open-ended research questions using data from 12 datasets and research questions drawn from existing scientific literature.

Copied to clipboard

Junru Song, Hengzhe Jin, Yucong Huang, Tingsong Jiang, Weien Zhou, Feifei Wang, Yang Yang, Ying Wen, Wen Yao

Agent-RewardBench: Towards a Unified Benchmark for Reward Modeling across Perception, Planning, and Safety in Real-World Multimodal Agents (2025.acl-long)

Challenge:	a lightweight world model converts raw pixels into object-centric symbolic states amenable to language-based reasoning . IMPLEMENT is a framework for grounding language agents in visual embodied environments .
Approach:	They propose a model-based reasoning framework that enables frozen large language models to perform imaginative planning.
Outcome:	The proposed framework can be used to ground language agents in visual embodied environments.

Copied to clipboard

Tianyi Men, Zhuoran Jin, Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao

Little Red Riding Hood Goes around the Globe: Crosslingual Story Planning and Generation with Large Language Models (2024.lrec-main)

Challenge:	Multimodal Large Language Models (MLLMs) are developing but lack external feedback . there is no clear on how to select reward models for agents .
Approach:	They propose a benchmark to evaluate agent reward modeling ability in MLLMs . they use multiple dimensions and real-world agent scenarios evaluation .
Outcome:	The proposed benchmark evaluates agent performance in multimodal large language models . it covers perception, planning, and safety with 7 scenarios and is highly difficult and high-quality .

Copied to clipboard

Evgeniia Razumovskaia, Joshua Maynez, Annie Louis, Mirella Lapata, Shashi Narayan

Multilinguality

Challenge:	Existing work has demonstrated the effectiveness of planning for story generation exclusively in a monolingual setting focusing primarily on English.
Approach:	They propose a task of crosslingual story generation with planning to leverage the creative and reasoning capabilities of large pretrained language models to generate stories in multiple languages.
Outcome:	The proposed task combines planning and planning in a monolingual setting and demonstrates that plans which structure stories into three acts lead to more coherent and interesting narratives while allowing to explicitly control their content and structure.

Explaining Mixtures of Sources in News Articles (2024.findings-emnlp)

Copied to clipboard

Alexander Spangher, James Youn, Matt DeButts, Nanyun Peng, Emilio Ferrara, Jonathan May

Explainability & Interpretability in NLP

Challenge:	a recent study shows that language models are essential for long-form article generation.
Approach:	They propose a generative process where a source-selection schema is first selected by a journalist, and then sources are chosen based on categories in that schema.
Outcome:	The proposed model can predict the most suitable schema given just the headline with reasonable accuracy.

Mathador-LM: A Dynamic Benchmark for Mathematical Reasoning on Large Language Models (2024.emnlp-main)

Copied to clipboard

Eldar Kurtic, Amir Moeini, Dan Alistarh

Interactive and Expressive Code-Augmented Planning with Large Language Models (2025.acl-long)

Challenge:	a new benchmark for evaluating the mathematical reasoning on large language models is being developed . popularity of reasoning benchmarks is leading to performance saturation and training set contamination.
Approach:	They introduce a benchmark for evaluating the mathematical reasoning on large language models . they find that models struggle with Mathador-LM, scoring lower than average 3rd graders .
Outcome:	The proposed benchmark improves performance on large language models . it also reduces test-set leakage into training data, a new study shows .

Copied to clipboard

Anthony Zhe Liu, Xinhe Wang, Jacob Sansom, Yao Fu, Jongwook Choi, Sungryull Sohn, Jaekyeom Kim, Honglak Lee

MoEC: A Memory-Routed Mixture-of-Experts Controller for Adaptive Minecraft Control (2026.acl-long)

Challenge:	Large Language Models (LLMs) have strong abilities in common-sense reasoning and interactive decision-making, but struggle with complex, long-horizon planning tasks.
Approach:	They propose a code-based LLM planning approach that is code-expressive while also dynamically adapting from errors.
Outcome:	The proposed approach can be error-prone and insufficient for handling ambiguous or unstructured data.

Copied to clipboard

Hui Wu, Chao Xu, Jianghui Wang, Ziqiong Liu, Dong Li, Yiwei Dai, Emad Barsoum

World Modeling Makes a Better Planner: Dual Preference Optimization for Embodied Task Planning (2025.acl-long)

Challenge:	Existing systems rely on a monolithic policy to execute subgoals across varying contexts, causing inconsistent outcomes and scaling only partially mitigates.
Approach:	They propose a memory-routed mixtureof-experts controller for Adaptive Minecraft Control that routes via a subgoal-indexed expert memory and regulates capacity through failure-triggered expert growth and redundancy-aware consolidation.
Outcome:	The proposed controller shows significant gains in adaptability, robustness, and execution consistency over strong baselines.

Copied to clipboard

Siyin Wang, Zhaoye Fei, Qinyuan Cheng, Shiduo Zhang, Panpan Cai, Jinlan Fu, Xipeng Qiu

Green & Sustainable NLP

Challenge:	Existing approaches focus on action selection or use pre-trained models as world models to enhance planning capabilities.
Approach:	They propose a new learning framework that optimizes state prediction and action selection through preference learning.
Outcome:	The proposed method outperforms existing methods and GPT-4o on VoTa-Bench and Qwen2-VL (7B), LLaVA-1.6 (7B) and LLama-3.2 (11B).

CONSENSAGENT: Towards Efficient and Effective Consensus in Multi-Agent LLM Interactions Through Sycophancy Mitigation (2025.findings-acl)

Copied to clipboard

Priya Pitre, Naren Ramakrishnan, Xuan Wang

TCP: a Benchmark for Temporal Constraint-Based Planning (2025.emnlp-main)

Challenge:	Recent advances in multi-agent large language model systems have shown remarkable performance in tasks such as reasoning, planning, and decision-making.
Approach:	They propose a framework that dynamically refines prompts based on agent interactions to mitigate sycophancy by requiring additional debate rounds to reach consensus.
Outcome:	The proposed framework outperforms both single-agent and multi-a agent baselines and achieves state-of-the-art results across all benchmark datasets.

Copied to clipboard

Zifeng Ding, Sikuan Yan, Moy Yuan, Xianglong Hu, Fangru Lin, Andreas Vlachos

Beyond the Answer: Advancing Multi-Hop QA with Fine-Grained Graph Reasoning and Evaluation (2025.acl-long)

Challenge:	Existing benchmarks evaluate temporal reasoning and planning in isolation and under limited forms of complexity.
Approach:	They propose a temporal constraint-based planning benchmark that assesses temporal reasoning and planning capabilities in large language models.
Outcome:	The proposed model fails to perform well under limited constraints and lacks temporal grounding.

Copied to clipboard

Qichuan Liu, Chentao Zhang, Chenfeng Zheng, Guosheng Hu, Xiaodong Li, Zhihong Zhang

Structured Data in NLP

Challenge:	Existing evaluations of multi-hop question answering systems focus on comparing final answers of reasoning method and given ground-truths.
Approach:	They propose a "Planner-Executor-Reasoner" architecture that evaluates reasoning . they propose PER-DP and PER QA architectures that provide ground-truths .
Outcome:	The proposed model improves the performance of multi-hop question answering systems.

A Category-Theoretic Approach to Neural-Symbolic Task Planning with Bidirectional Search (2025.findings-emnlp)

Copied to clipboard

Shuhui Qu, Jie Wang, Kincho Law

GUI Agents: A Survey (2025.findings-acl)

Challenge:	Empirical evaluations demonstrate that our method improves completion rates by up to 6.6% and action accuracy by 9.1% .
Approach:	They propose a Neural-Symbolic Task Planning framework that integrates Large Language Model (LLM) decomposition with category-theoretic verification for resource-aware, temporally consistent planning.
Outcome:	The proposed framework improves completion rates and action accuracy by up to 6.6% . it also eliminates resource violations while ensuring resource-awareness and consistency.

Copied to clipboard

Dang Nguyen, Jian Chen, Yu Wang, Gang Wu, Namyong Park, Zhengmian Hu, Hanjia Lyu, Junda Wu, Ryan Aponte, Yu Xia, Xintong Li, Jing Shi, Hongjie Chen, Viet Dac Lai, Zhouhang Xie, Sungchul Kim, Ruiyi Zhang, Tong Yu, Mehrab Tanjim, Nesreen K. Ahmed, Puneet Mathur, Seunghyun Yoon, Lina Yao, Branislav Kveton, Jihyung Kil, Thien Huu Nguyen, Trung Bui, Tianyi Zhou, Ryan A. Rossi, Franck Dernoncourt

Challenge:	Large Foundation Models (LFMs) have transformed the landscape of AI research and day-to-day life.
Approach:	They propose a framework that delineates GUI agents' perception, reasoning, planning, and acting capabilities.
Outcome:	The proposed framework delineates their perception, reasoning, planning, and acting capabilities.

PRINCIPLES: Synthetic Strategy Memory for Proactive Dialogue Agents (2025.findings-emnlp)

Copied to clipboard

Namyoung Kim, Kai Tzu-iunn Ong, Yeonjun Hwang, Minseok Kang, Iiseo Jihn, Gayoung Kim, Minju Kim, Jinyoung Yeo

Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks (2025.acl-long)

Challenge:	Existing strategies for proactive dialogue face limitations such as limited strategy coverage and preference bias in planning.
Approach:	They propose a synthetic strategy memory for proactive dialogue agents based on large language models . PRINCIPLES is derived through offline self-play simulations and serves as reusable knowledge that guides strategy planning during inference.
Outcome:	PRINCIPLES is a synthetic strategy memory for proactive dialogue agents.

Copied to clipboard

Xingxuan Li, Weiwen Xu, Ruochen Zhao, Fangkai Jiao, Shafiq Joty, Lidong Bing

Programming Languages in NLP

Challenge:	Existing approaches to problem-solving for large language models fail to provide accurate reasoning and factual accuracy.
Approach:	They propose a framework that leverages fine-tuned critic models to guide reasoning and retrieval processes.
Outcome:	The proposed framework outperforms baselines on domain-knowledge-intensive tasks . it can be used to iterate retrieval and reasoning, and improve retrieval relevance .

ATLAS: Agent Tuning via Learning Critical Steps (2025.findings-acl)

Copied to clipboard

Zhixun Chen, Ming Li, Yuxuan Huang, Yali Du, Meng Fang, Tianyi Zhou

A Survey on Evaluation of LLM-based Agents (2026.findings-acl)

Challenge:	Existing agent tuning approaches employ supervised finetuning on entire expert trajectories, but behavior-cloning of full traitories introduces expert bias and weakens generalization to states not covered by the expert data.
Approach:	They propose a method that finetunes LLMs on critical steps in expert trajectories and identifies and finetuns them on these steps with reduced costs.
Outcome:	The proposed method outperforms existing methods and open-source LLM agents on only 30% critical steps in extensive experiments.

Copied to clipboard

Asaf Yehudai, Lilach Eden, Alan Li, Guy Uziel, Yilun Zhao, Roy Bar-Haim, Arman Cohan, Michal Shmueli-Scheuer

Challenge:	This paper provides the first comprehensive survey of evaluation methods for LLM-based agents . LLMs are static, having fixed knowledge, and confined to text-to-text interaction.
Approach:	They analyze the evaluation of LLM-based agents across five perspectives . they identify current trends and key gaps in evaluation methods .
Outcome:	The proposed evaluation frameworks and tools are based on five perspectives . the results highlight current trends and identify gaps in future research .

STA-CoT: Structured Target-Centric Agentic Chain-of-Thought for Consistent Multi-Image Geological Reasoning (2025.findings-emnlp)

Copied to clipboard

Beibei Yu, Tao Shen, Ling Chen

TALON: A Multi-Agent Framework for Long-Table Exploration and Question Answering (2025.emnlp-main)

Challenge:	Reliable multi-image geological reasoning is essential for automating expert tasks in remote-sensing mineral exploration.
Approach:	They propose a framework that orchestrates planning, execution, and verification agents to decompose, ground, and iteratively refine reasoning steps over geological and hyperspectral image sets.
Outcome:	The proposed framework decomposes, ground, and iteratively refines reasoning steps over geological and hyperspectral image sets.

Copied to clipboard

Ruochun Jin, Xiyue Wang, Dong Wang, Haoqi Zheng, Yunpeng Qi, Silin Yang, Meng Zhang

MavenCoder: Competitive Code Generation via Model Adaptive Planning Strategies and Multi-Perspective Verification Enhancement (2026.acl-long)

Challenge:	Existing approaches to query-relevant content retrieval fail to retrieve contextually relevant data.
Approach:	They propose a multi-agent framework for table question answering over long tables . TALON features a planning agent that iteratively invokes a tool agent to access tabular data .
Outcome:	The proposed framework achieves average accuracy improvements of 7.5% and 12.0% across all language models.

Copied to clipboard

ZhenChun Xu, Yi Cai, Dajun Zheng, li Yuan, Mengchen Zhao, Qixiang Wang, Jiexin Wang

Programming Languages in NLP

Challenge:	Recent advances in large language models (LLMs) have significantly enhanced automated program synthesis.
Approach:	They propose a model-adaptive and verification–enhanced framework for competition-level code generation that leverages adaptive assessment aligned with the model’s capabilities to select planning strategies while providing timely feedback and correction via multi-perspective verification.
Outcome:	The proposed framework outperforms existing state-of-the-art approaches on livecodebench, humanEval+, MBPP+, and codecontests, and achieves pass@1 results exceeding 3%–40%.

TAMAS: Benchmarking Adversarial Risks in Multi-Agent LLM Systems (2026.acl-long)

Copied to clipboard

Ishan Kavathekar, Hemang Jain, Ameya Rathod, Ponnurangam Kumaraguru, Tanuja Ganu

RSMeM: Knowledge-Enhanced Memory Evolution for Remote Sensing Agents with Systematic Evaluation (2026.acl-long)

Challenge:	Existing benchmarks and datasets focus on single-agent settings, failing to capture the unique vulnerabilities of multi-agend LLM dynamics and co-ordination.
Approach:	They propose a benchmark to evaluate the robustness and safety of multi-agent LLM systems.
Outcome:	The proposed benchmark evaluates the robustness and safety of multi-agent LLM systems.

Copied to clipboard

Bingxian Wu, Yu Zhang, Zonghao Guo, Tang Liu, Chen Qian, Yuxiang Lu, Xingbo Du, Yanghao Li, Yidan Zhang, Chi Chen, Ling Yao, Chenghu Zhou, Maosong Sun

MA2P: A Meta-Cognitive Autonomous Intelligent Agents Framework for Complex Persuasion (2026.findings-acl)

Challenge:	Existing RS agents built on general-purpose LLMs are domain-agnostic, resulting in brittle and error-prone workflows.
Approach:	They propose a knowledge-enhanced memory evolution mechanism that bootstraps RS agents with pre-distilled domain knowledge and iteratively integrates online experience for robust multi-step tool execution.
Outcome:	Experiments show that the new model improves tool-use performance and accuracy . iteratively, iteration of the model integrates online experience for robust multi-step tool execution .

Copied to clipboard

Dingyi Zhang, Ziqing Zhuang, Linhai Zhang, Ziyang Gao, Deyu Zhou

Prior Prompt Engineering for Reinforcement Fine-Tuning (2025.emnlp-main)

Challenge:	Existing approaches to persuasion generate generic or weakly grounded responses even when such cues are identified.
Approach:	They propose a meta-cognitive autonomous intelligent agent framework for complex persuasion that coordinates perception management, mental-state inference, strategy execution, memory maintenance, and performance evaluation.
Outcome:	The proposed framework achieves a higher persuasion success rate than baselines.

Copied to clipboard

Pittawat Taveekitworachai, Potsawee Manakul, Sarana Nutanong, Kunat Pipatanakul

GAVEL: Evidence-Contract Debate with Mechanized Scrutiny for Provenance-Grounded Fact-Checking (2026.findings-acl)

Challenge:	Existing studies have focused on algorithms, reward shaping, and data curation, but prior prompt engineering is understudied.
Approach:	They investigate prior prompt engineering (pPE) in reinforcement fine-tuning . they translate five representative iPE strategies into corresponding pPE approaches .
Outcome:	The proposed approaches outperform iPE-prompted models on in-domain and out-of-domain benchmarks.

Copied to clipboard

Ruoyu Xu, Gaoxiang Li, Victor S. Sheng

LongVideoAgent: Multi-Agent Reasoning with Long Videos (2026.acl-long)

Challenge:	Evidence-grounded fact-checking requires predicting claim veracity while returning faithful evidence at fine granularity.
Approach:	They propose a multi-agent debate framework that enforces evidence grounding throughout inference.
Outcome:	The proposed framework improves provenance-aware metrics over existing frameworks.

Copied to clipboard

Runtao Liu, Ziyi Liu, Jiaqi Tang, Yue Ma, Renjie Pi, Jipeng Zhang, Qifeng Chen

Embodied-Reasoner: Synergizing Visual Search, Reasoning, and Action for Embodied Interactive Tasks (2026.acl-long)

Challenge:	a key emerging challenge is robust long video understanding, authors say . current methods compress content into lossy summaries or rely on limited toolsets .
Approach:	They propose a multi-agent framework where a master LLM coordinates a grounding agent and a vision agent to extract targeted textual observations.
Outcome:	The proposed model outperforms strong non-agent baselines on episode-level datasets . the proposed model significantly outperformed existing models on other datasets.

Copied to clipboard

Wenqi Zhang, Mengna Wang, Gangao Liu, Huixin Xu, Yiwei Jiang, Yongliang Shen, Guiyang Hou, Zhe Zheng, Hang Zhang, Xin Li, Jiajun Liu, Weiming Lu, Peng Li, Yueting Zhuang

Characterizing the Robustness of Black-Box LLM Planners Under Perturbed Observations with Adaptive Stress Testing (2026.findings-acl)

Challenge:	Recent advances in reasoning models have demonstrated remarkable capabilities on mathematical and coding tasks, but their effectiveness in embodied domains remains largely unexplored.
Approach:	They propose a reasoning model for interactive embodied tasks that synthesizes 9.3k coherent Observation-Thought-Action trajectories containing 64k ego-centric images and 90k diverse reasoning processes.
Outcome:	The proposed model outperforms existing visual reasoning models by +9%, 24%, and +13% on long-horizon tasks.

Copied to clipboard

Neeloy Chakraborty, John Pohovey, Melkior Ornik, Katherine Rose Driggs-Campbell

AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments (2026.acl-long)

Challenge:	Large language models have been used in reasoning-heavy planning tasks but their tendency to hallucinate unsafe outputs poses risks.
Approach:	They propose a method for searching the space of prompt perturbations using adaptive stress testing with Monte-Carlo tree search.
Outcome:	The proposed method can detect scenarios, sensor configurations, and prompt phrasing that cause models to act with high uncertainty or crash.

Copied to clipboard

Zhiheng Xi, Dingwen Yang, Jiaqi Liu, Jixuan Huang, Honglin Guo, Baodai Huang, Tinggang Chen, Qi Zhang, Zhonghang Lu, Chenyu Liu, Jiajun Sun, Jiazheng Zhang, Dingwei Zhu, Xin Guo, Junzhe Wang, Zhihao Zhang, Yuming Yang, Junjie Ye, Minghe Gao, Dongrui Liu, Jiaming Ji, Guohao Li, Tao Gui, Qi Zhang, Xuanjing Huang

Revealing the Seen, Imagining the Beyond: A Survey of Image-Grounded Chain-of-Thought Reasoning in Multimodal LLMs (2026.acl-long)

Challenge:	Existing benchmarks evaluate agents in simplified, idealized settings, relying on pre-packaged tool interfaces, overlooking critical steps, and assume inputs are clean and fully specified.
Approach:	They propose a framework that evaluates language agents in simplified, idealized settings . they show that even SOTA systems like Gemini and GPT-5 struggle on AgentGym2 .
Outcome:	Experiments on 15 proprietary and open-source models show that even SOTA systems like Gemini and GPT-5 struggle on AgentGym2 .

Copied to clipboard

Qihua Dong, Yitian Zhang, Huimin Zeng, Yizhou Wang, Jianglin Lu, Kuo Yang, Yun Fu

Internalizing Multi-Agent Reasoning for Accurate and Efficient LLM-based Recommendation (2026.findings-acl)

Challenge:	Recent advances in Multimodal Large Language Models (MLLMs) have shifted visual reasoning from tool-calling to end-to-end perceptionreasoning.
Approach:	They synthesize the emerging paradigm of Image-Grounded Chain-of-Thought (IG-CoT) they propose a method-centric taxonomy covering prompting, supervised fine-tuning, and reinforcement learning .
Outcome:	The proposed model is based on a method-centric taxonomy and benchmarks.

Copied to clipboard

Yang Wu, Haoze Wang, Qian Li, Jun Zhang, Huan Yu, Jie Jiang