REASONING ACCURACY

EquivPruner: Boosting Efficiency and Quality in LLM-Based Search via Action Pruning (2026.acl-long)

Challenge:	Large Language Models (LLMs) exhibit impressive performance across various domains but struggle with arithmetic reasoning tasks.
Approach:	They propose a Teaching-Inspired Integrated Prompting Framework which emulates the instructional process of a teacher guiding students.
Outcome:	The proposed framework improves reasoning accuracy on nine benchmarks.

Copied to clipboard

Jiawei Liu, Qisi Chen, Jianshu Zhang, Quan Liu, Defu Lian

Can Activation Steering Generalize Across Languages? A Study on Syllogistic Reasoning in Language Models (2026.eacl-long)

Challenge:	Existing semantic similarity methods struggle to accurately identify semantically equivalent steps in domain-specific contexts like mathematical reasoning.
Approach:	They propose a simple yet effective approach that identifies and prunes semantically equivalent actions during LLM reasoning search.
Outcome:	The proposed approach reduces token consumption while improving reasoning efficiency and accuracy on Qwen2.5-Math-7B-Instruct on GSM8K.

Copied to clipboard

Gabriele Maraia, Leonardo Ranaldi, Marco Valentino, Fabio Massimo Zanzotto

Mirror-Consistency: Harnessing Inconsistency in Majority Voting (2024.findings-emnlp)

Challenge:	Prior work has focused on activation steering for Large Language Models (LLMs) this technique can be used to improve reasoning accuracy and transferability across languages.
Approach:	They propose to use activation steering to steer models towards a cross-lingual reasoning space.
Outcome:	The proposed techniques generalise well to multilingual datasets while minimizing language modelling performance.

Copied to clipboard

Siyuan Huang, Zhiyuan Ma, Jintao Du, Changhua Meng, Weiqiang Wang, Zhouhan Lin

Unleashing the Reasoning Potential of LLMs by Critique Fine-Tuning on One Problem (2025.emnlp-main)

Challenge:	Large Language Models (LLMs) are a widely-used decoding strategy that relies on the plurality voting rule, which focuses on the most frequent answer while overlooking all other minority responses.
Approach:	They propose to incorporate a ‘reflective mirror’ into the self-ensemble decoding process and enables LLMs to critically examine inconsistencies among multiple generations.
Outcome:	The proposed method incorporates a ‘reflective mirror’ into the self-ensemble decoding process and enables LLMs to critically examine inconsistencies among multiple generations.

Copied to clipboard

Yubo Wang, Ping Nie, Kai Zou, Lijun Wu, Wenhu Chen

CAT: Confidence-Adaptive Thinking for Efficient Reasoning of Large Reasoning Models (2026.acl-industry)

Challenge:	Critique Fine-Tuning (CFT) is a promising paradigm for unlocking the reasoning capabilities of large language models.
Approach:	They propose a method that leverages critique data generated from a single math problem to improve reasoning accuracy.
Outcome:	The proposed method surpasses one-shot RLVR while requiring 15 to 20 times less compute.

Copied to clipboard

Qizhi Jiang, Shuo Wang, Pei Ke, Yuhang Song, Ke Qin

Cross-lingual Prompting: Improving Zero-shot Chain-of-Thought Reasoning across Languages (2023.emnlp-main)

Challenge:	Existing compression methods for large reasoning models rely on uniform length reduction or coarse-grained difficulty estimation, often leading to performance degradation on difficult problems.
Approach:	They propose a framework that incorporates model’s intrinsic self-certainty signals as confidence into the preference optimization process, which autonomously modulates reasoning lengths based on problem difficulty.
Outcome:	The proposed framework outperforms state-of-the-art models on reasoning accuracy across multiple benchmarks on different base models.

Copied to clipboard

Libo Qin, Qiguang Chen, Fuxuan Wei, Shijue Huang, Wanxiang Che

Low-Resource NLP

Challenge:	Existing methods for zero-shot CoT are limited to a single language, making it difficult to generalize to other languages and hindering global development.
Approach:	They introduce cross-lingual prompting (CLP) to improve zero-shot CoT reasoning across languages.
Outcome:	The proposed method outperforms existing prompting methods on several benchmarks.

DRIVINGVQA: A Dataset for Interleaved Visual Chain-of-Thought in Real-World Driving Scenarios (2026.findings-eacl)

Copied to clipboard

Charles Corbière, Simon Roburin, Syrielle Montariol, Antoine Bosselut, Alexandre Alahi

JUREX-4E: Juridical Expert-Annotated Four-Element Knowledge Base for Legal Reasoning (2025.emnlp-main)

Challenge:	Chain-of-thought (CoT) prompting is a prompting strategy that improves reasoning in large language models, but its effectiveness in vision-language models remains limited due to over-reliance on textual cues and memorized knowledge.
Approach:	They propose a visual question-answering dataset derived from driving theory exams that incorporates textual explanations with visual tokens extracted from entities relevant to the reasoning process.
Outcome:	The proposed approach outperforms chain-of-thought prompting in large language models and vision-language models in real-world scenarios.

Copied to clipboard

Huanghai Liu, Quzhe Huang, Qingjing Chen, Yiran Hu, Jiayu Ma, Yun Liu, Weixing Shen, Yansong Feng

Knowledge Representation

Challenge:	Recent studies have introduced legal theories into LLM workflows to improve their understanding of legal texts and reasoning accuracy.
Approach:	They evaluate an expert-annotated four-element knowledge base covering 155 criminal charges.
Outcome:	The proposed model can be used to analyze criminal charges and retrieve them in legal cases.

Entropy-based Exploration Conduction for Multi-step Reasoning (2025.findings-acl)

Copied to clipboard

Jinghan Zhang, Xiting Wang, Fengran Mo, Yeyang Zhou, Wanfu Gao, Kunpeng Liu

Can Knowledge Graphs Reduce Hallucinations in LLMs? : A Survey (2024.naacl-long)

Challenge:	Existing methods to automatically decide the depth of exploration of the reasoning procedure lead to high cost and a lack of flexibility.
Approach:	They propose a method that dynamically adjusts the exploration depth during multi-step reasoning by monitoring LLM’s output entropy and variance entropic.
Outcome:	The proposed method captures the uncertainty of the current step and the fluctuation of uncertainty across consecutive reasoning steps and then selects whether to deepen, expand, or stop exploration according to the probability.

Copied to clipboard

Garima Agrawal, Tharindu Kumarage, Zeyad Alghamdi, Huan Liu

Structured Data in NLP

Challenge:	Increasing the use of knowledge graphs to augment LLMs has led to hallucinations . large language models (LLMs) are prone to producing hallucinosis due to knowledge gaps .
Approach:	They review knowledge graph-based augmentation techniques in large language models to assess their effectiveness and examine their performance.
Outcome:	The proposed methods have been evaluated against three groups of LLMs and offer methodological comparisons and performance evaluations.

ChartMind: A Comprehensive Benchmark for Complex Real-world Multimodal Chart Question Answering (2025.emnlp-main)

Copied to clipboard

Jingxuan Wei, Nan Xu, Junnan Zhu, null Haoyanni, Gaowei Wu, Qi Chen, Bihui Yu, Lei Wang

Natural Language Interfaces

Challenge:	Chart question answering (CQA) is a multimodal task for evaluating the reasoning capabilities of vision-language models.
Approach:	They propose a chart question answering benchmark that incorporates multilingual contexts and supports open-domain textual outputs.
Outcome:	The proposed framework outperforms the previous three common CQA paradigms: instruction-following, OCR-enhanced, and chain-of-thought.

MetaLadder: Ascending Mathematical Solution Quality via Analogical-Problem Reasoning Transfer (2025.findings-emnlp)

Copied to clipboard

Honglin Lin, Zhuoshi Pan, Qizhi Pei, Xin Gao, Yu Li, Mengzhang Cai, Conghui He, Lijun Wu

GRPO-LEAD: A Difficulty-Aware Reinforcement Learning Approach for Concise Mathematical Reasoning in Language Models (2025.emnlp-main)

Challenge:	Current paradigms generate CoT and answers directly for a given problem, diverging from human problem-solving strategies to some extent.
Approach:	They propose a framework that explicitly prompts LLMs to recall and reflect on meta-problems alongside their CoT solutions before addressing the target problem.
Outcome:	The proposed framework outperforms standard CoT-based methods on mathematical benchmarks and significantly improves their reasoning accuracy.

Copied to clipboard

Jixiao Zhang, Chunsheng Zuo

Think Hard Only When Needed: A Hybrid Best-of-N and Beam Search for Efficient Test-Time Compute (2026.findings-eacl)

Challenge:	Existing methods for group-relative policy optimization face challenges in reward sparsity, verbosity and inadequate focus on problem difficulty.
Approach:	They propose a method to improve group relative policy optimization with length-regularized rewards and explicit penalties for incorrect solutions.
Outcome:	The proposed method achieves state-of-the-art performance for 14B-scale models . it improves reasoning accuracy, conciseness, and efficiency .

Copied to clipboard

Hyewon Suh, Chaojian Li, Cheng-Jhih Shih, Zheng Wang, Kejing Xia, Yonggan Fu, Yingyan Celine Lin

Neuro-Symbolic Integration Brings Causal and Reliable Reasoning Proofs (2025.findings-naacl)

Challenge:	Large language models (LLMs) exhibit remarkable reasoning and planning capabilities, yet their substantial inference-time cost significantly impedes deployment in resourceconstrained applications.
Approach:	They propose a hybrid inference pipeline that combines beam search and Best-of-N . THROW generates shorter initial trajectories and evaluates them using PRMs .
Outcome:	THROW achieves 1.54 and 14.38 latency speedups and 35.7% and 80.4% token reductions on average compared to Best-of-N and beam search .

Copied to clipboard

Sen Yang, Xin Li, Leyang Cui, Lidong Bing, Wai Lam

Explainability & Interpretability in NLP

Challenge:	a new framework for complex reasoning with LLMs is developed to improve reasoning proof accuracy and interpretability.
Approach:	They propose to use LLMs to generate search logs that can be interpreted into human-readable reasoning proofs.
Outcome:	The proposed framework improves reasoning accuracy but lacks interpretability due to black-box nature of the solvers.

The Box is in the Pen: Evaluating Commonsense Reasoning in Neural Machine Translation (2020.findings-emnlp)

Copied to clipboard

Jie He, Tao Wang, Deyi Xiong, Qun Liu

Text Generation

Challenge:	a test suite to evaluate commonsense reasoning capability of neural machine translation is presented . language models pretrained on large-scale corpora achieve a commonsensing accuracy of lower than 72% on target translations of this test suite.
Approach:	They propose a test suite to evaluate the commonsense reasoning capability of neural machine translation.
Outcome:	The proposed test suite performs poorly on commonsense reasoning of the three ambiguity types in terms of reasoning accuracy and reasoning consistency.

How Language Models Conflate Logical Validity with Plausibility: A Representational Analysis of Content Effects (2026.findings-acl)

Copied to clipboard

Leonardo Bertolazzi, Sandro Pezzelle, Raffaella Bernardi

Semantic Text Processing

Challenge:	a number of theories have been proposed to account for content effects in large language models, including the dual-process theory of reasoning, but the mechanisms behind content effects remain unclear.
Approach:	They propose to encode validity and plausibility concepts in LLMs by aligning them in representational geometry.
Outcome:	The proposed model conflates validity and plausibility, and vice versa.

EffiQA: Efficient Question-Answering with Strategic Multi-Model Collaboration on Knowledge Graphs (2025.coling-main)

Copied to clipboard

Zixuan Dong, Baoyun Peng, Yufei Wang, Jia Fu, Xiaodong Wang, Xin Zhou, Yongxue Shan, Kangchen Zhu, Weiguo Chen

Structured Data in NLP

Challenge:	Existing approaches that integrate LLMs and KGs either underutilize the reasoning abilities of LLM or suffer from prohibitive computational costs due to tight coupling.
Approach:	They propose a framework that can strike a balance between performance and efficiency via an iterative paradigm.
Outcome:	The proposed framework can strike a balance between performance and efficiency via an iterative paradigm.

Learning to Correct for QA Reasoning with Black-box LLMs (2024.emnlp-main)

Copied to clipboard

Jaehyung Kim, Dongyoung Kim, Yiming Yang

Chain-of-Memory: Lightweight Memory Construction with Dynamic Evolution for LLM Agents (2026.acl-long)

Challenge:	Existing approaches to improve reasoning capability of large language models rely on accessibility or require significantly increased train- and inference-time costs.
Approach:	They propose a method to improve QA reasoning of large language models in a black-box setting by using a trained adaptation model to perform a seq2seq mapping from the often-imperfect reasonings of the original LLM to the correct or improved reasonings.
Outcome:	The proposed approach significantly improves reasoning accuracy across various QA benchmarks compared to the best-performing adaptation baselines.

Copied to clipboard

Xiucheng Xu, Bingbing Xu, Tian Xueyun, Zihe Huang, Rongxin Chen, Li Yunfan, Huawei Shen

Information Retrieval

Challenge:	Existing paradigms for large language model (LLM) agents use memory construction and retrieval-augmented generation.
Approach:	They propose a framework that advocates for a paradigm shift toward lightweight construction paired with sophisticated utilization.
Outcome:	Experiments show that CoM outperforms baselines with accuracy gains of 7.5%–10.4% while reducing computational overhead to approximately 2.7% of token consumption and 6.0% of latency compared to complex memory architectures.

DS-MHP: Improving Chain-of-Thought through Dynamic Subgraph-Guided Multi-Hop Path (2025.findings-emnlp)

Copied to clipboard

Yongqiang Liu, Qiyao Peng, Binrong Liu, Hongtao Liu, XueWei Li, Wenjun Wang

RoboFailRing: Retrieval-Augmented and Language Grounding Failure Detection for VLM-enabled Robotic Manipulation (2026.acl-long)

Challenge:	Existing knowledge graph methods lack adaptability in knowledge-intensive tasks with multiple entities and implicit multi-hop relations.
Approach:	They propose a zero-shot framework to enhance LLM reasoning in multi-entity relation tasks.
Outcome:	DS-MHP outperforms baselines and state-of-the-art methods on 12 datasets spanning commonsense, logical, symbolic, and arithmetic reasoning.

Copied to clipboard

Chenduo Ying, Linkang Du, Yuanchao Shu, Peng Cheng

Information Retrieval

Challenge:	RoboFailRing enables timely failure detection during task execution and enhances reasoning accuracy of VLMs.
Approach:	They propose a robot-based failure detection system that enables timely failure detection . they evaluate a large-scale simulated dataset and provide a grounded failure report .
Outcome:	The proposed method achieves rapid failure detection and returns similarity-based decision on large-scale simulated failures.

R2I-Bench: Benchmarking Reasoning-Driven Text-to-Image Generation (2025.emnlp-main)

Copied to clipboard

Kaijie Chen, Zihao Lin, Zhiyang Xu, Ying Shen, Yuguang Yao, Joy Rimchala, Jiaxin Zhang, Lifu Huang

Revisiting Pruning vs Quantization for Small Language Models (2025.findings-emnlp)

Challenge:	Reasoning is a fundamental capability underpinning text-to-image (T2I) generation.
Approach:	They propose a benchmark to rigorously assess reasoning-driven T2I generation.
Outcome:	Experiments with 16 representative T2I models show limited reasoning performance . a strong pipeline-based framework decouples reasoning and generation .

Copied to clipboard

Zihan Zhou, Simon Kurz, Zhixue Zhao

DisCal: Distribution-Aware Calibration for Mathematical Reasoning Under Character-Level Noisy Inputs (2026.acl-long)

Challenge:	Compressing Small Language Models (SLMs) is particularly suited for resource-constrained devices, but their compression dynamics remain underexplored compared to Large Language Model (LLMs).
Approach:	They evaluated post-training pruning and quantization methods across six SLMs from 0.5 to 3.8B, seven languages, and seven downstream tasks.
Outcome:	The proposed methods outperform pruning and quantization on six SLMs from 0.5 to 3.8B, seven languages, and seven downstream tasks.

Copied to clipboard

Bo Zhang, Jiawei Zhang, Cong Gao, Bingxu Han, Minghao Hu, Jun Zhang, Yunbo Cao, Zhunchen Luo, Wen Yao, Guotong Geng, Zhong Wang

QueryLink: Leveraging Query-Memory Alignment for Long-Term Reasoning in LLM Agents (2026.findings-acl)

Challenge:	Existing methods for calibration of large reasoning models (LRMs) focus on clean inputs, leaving noise unexplored.
Approach:	They propose a confidence calibration framework for character-level noisy inputs that extracts uncertainty signals from both the empirical answer distribution and the model’s predictive distribution and integrates them via a learned calibrator.
Outcome:	Experiments on multiple mathematical reasoning benchmarks show that DisCal outperforms existing calibration methods under noisy inputs, reducing expected calibration error (ECE) by up to 39.21% and improving Area Under the Receiver Operating Characteristic Curve (AUROC) by 31.44%.

Copied to clipboard

Xuxian Hu, Zhu Teng, Wei Zhang, Ming He, Jianping Fan

Information Retrieval

Challenge:	Existing approaches to integrating external memory prioritize memory organization while overlooking a critical semantic gap between implicit, intent-driven queries and explicit, narrative-based memories.
Approach:	They propose a framework that leverages Query-Memory Alignment to project both queries and memories into a shared semantic space.
Outcome:	The proposed framework significantly outperforms SOTA methods on the LoCoMo and LongMemEval benchmarks and can be integrated as a plug-and-play component to boost existing vector-based systems like A-MEM.

ReCUT: Balancing Reasoning Length and Accuracy in LLMs via Stepwise Trails and Preference Optimization (2025.findings-emnlp)

Copied to clipboard

Zhensheng Jin, Xinze Li, Yifan Ji, Chunyi Peng, Zhenghao Liu, Qi Shi, Yukun Yan, Shuo Wang, Furong Peng, Ge Yu

RISE: Reasoning Enhancement via Iterative Self-Exploration in Multi-hop Question Answering (2025.findings-acl)

Challenge:	Existing methods to train LLMs suffer from overthinking, leading to lengthy reasoning traces . Existing approaches to train large language models suffer from this problem .
Approach:	They propose a method to combine multiple reasoning chains for training LLMs . they use stepwise exploration and long-short switched sampling to evaluate reasoning paths .
Outcome:	The proposed method reduces reasoning lengths by approximately 30-50% . it also maintains or improves reasoning accuracy compared to baselines .

Copied to clipboard

Bolei He, Xinran He, Mengke Chen, Xianwei Xue, Ying Zhu, Zhen-Hua Ling

Natural Language Interfaces

Challenge:	Large Language Models (LLMs) excel in many areas but face challenges with complex reasoning tasks, such as Multi-Hop Question Answering (MHQA).
Approach:	They propose a framework to enhance models’ reasoning capability through iterative self-exploration that addresses key errors in MHQA tasks such as Evidence Aggregation and Reasoning Decomposition.
Outcome:	Extensive experiments on multiple MHQA benchmarks show that the proposed framework significantly improves reasoning accuracy and task performance.

Nash CoT: Multi-Path Inference with Preference Equilibrium (2024.emnlp-main)

Copied to clipboard

Ziqi Zhang, Cunxiang Wang, Xiao Xiong, Yue Zhang, Donglin Wang

Local Look-Ahead Guidance via Verifier-in-the-Loop for Automated Theorem Proving (2025.findings-acl)

Challenge:	Multi-path inference is an improvement on multi-path reasoning, but there is no optimal setting for the number of inference paths.
Approach:	They propose to use question-related role templates to guide LLMs into relevant roles to reduce the dependence on the number of inference paths.
Outcome:	The proposed system can achieve comparable or better results than self-consistency with the same number of paths.

Copied to clipboard

Sara Rajaee, Kumar Pratik, Gabriele Cesa, Arash Behboodi

Calibration-Aware Policy Optimization for Reasoning LLMs (2026.acl-long)

Challenge:	Recent methods for AI reasoning require applying variants of reinforcement learning (RL) on rolled out trajectories, even for step-wise rewards, or large quantities of human-annotated trajectory data.
Approach:	They propose a verifier-in-the-loop design that uses an automated verifier to give intermediate feedback at each step of the reasoning process.
Outcome:	The proposed model improves on the Automatic Theorem Proving task using Lean as the verifier.

Copied to clipboard

Ziqi Wang, Xingzhou Lou, Meiqi Wu, Zhengqi Wen, Junge Zhang

Stepwise Reasoning Checkpoint Analysis: A Test Time Scaling Method to Enhance LLMs’ Reasoning (2025.emnlp-main)

Challenge:	Existing approaches to model calibration are limited or sacrifice gains in reasoning accuracy.
Approach:	They propose a method that improves calibration by 15% while boosting accuracy by 5% . they propose GRPO-style algorithms that misalign uncertainty-agnostic advantage estimation .
Outcome:	The proposed approach improves calibration by 15% while achieving comparable to or better than GRPO on multiple mathematical reasoning benchmarks.

Copied to clipboard

Zezhong Wang, Xingshan Zeng, Weiwen Liu, Yufei Wang, Liangyou Li, Yasheng Wang, Lifeng Shang, Xin Jiang, Qun Liu, Kam-Fai Wong

VerifiAgent: a Unified Verification Agent in Language Model Reasoning (2025.findings-emnlp)

Challenge:	Existing methods that use Chain-of-Thought suffer from path homogenization and inefficient use of intermediate results.
Approach:	They propose a framework that introduces checkpoints between reasoning steps to reduce path homogenization and create fault-tolerant mechanisms.
Outcome:	The proposed framework reduces path homogenization and creates fault-tolerant mechanism by utilizing high-quality intermediate results.

Copied to clipboard

Jiuzhou Han, Wray Buntine, Ehsan Shareghi

Revisiting Entropy Regularization: Adaptive Coefficient Unlocks Its Potential for LLM Reinforcement Learning (2026.findings-acl)

Challenge:	Existing verification methods are typically model-specific or domain-restricted . Existing methods require significant computational resources and lack scalability .
Approach:	a unified verification agent integrates two levels of verification: meta-verification and tool-based adaptive verification.
Outcome:	The proposed agent outperforms baseline verification methods among reasoning tasks.

Copied to clipboard

Xiaoyun Zhang, Xiaojian Yuan, Di Huang, Wang You, Chen Hu, Jingqing Ruan, Kejiang Chen, Xing Hu

RARE: Retrieval-Augmented Reasoning Enhancement for Large Language Models (2025.acl-long)

Challenge:	Reasoning ability is a defining capability of Large Language Models (LLMs), but RLVR training suffers from policy entropy collapse, hindering exploration and limiting reasoning performance.
Approach:	They propose a framework that dynamically balances exploration and exploitation via three components: difficulty-aware coefficient allocation, initial-anchored target entropy, and dynamic global coefficient adjustment.
Outcome:	The proposed framework outperforms baselines on multiple mathematical reasoning benchmarks.

Copied to clipboard

Hieu Tran, Zonghai Yao, Zhichao Yang, Junda Wang, Yifan Zhang, Shuo Han, null Feiyun Ouyang, Hong Yu

Semantic Text Processing

Challenge:	Existing work aims to improve reasoning accuracy and factual integrity across large language models for knowledge-intensive tasks such as medical and commonsense reasoning.
Approach:	They propose a versatile extension to the mutual reasoning framework (rStar) that enhances reasoning accuracy and factual integrity across large language models.
Outcome:	The proposed extension to the mutual reasoning framework improves reasoning accuracy and factual integrity across large language models for complex, knowledge-intensive tasks.

Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention (2026.acl-long)

Copied to clipboard

Zhen Yang, Mingyang Zhang, Feng Chen, Ganggui Ding, Liang Hou, Xin Tao, Ying-Cong Chen

Leibniz: Theory-of-Mind Driven Neuro-Symbolic Logical Reasoning via Multi-Agent Collaboration (2026.acl-long)

Challenge:	Recent advances in large language models (LLMs) have focused on test-time scaling to improve reasoning quality but at the cost of efficiency.
Approach:	They propose a training-free framework that enhances reasoning accuracy and stability with minimal overhead.
Outcome:	The proposed framework yields consistent gains across general, coding, and STEM tasks while remaining highly efficient.

Copied to clipboard

Yue Fan, Hu Zhang, Yunxiao Zhao, Guangjun Zhang, Hao Zhan, Ru Li, Hongye Tan, Yuanlong Wang

Reversal of Thought: Enhancing Large Language Models with Preference-Guided Reverse Reasoning Warm-up (2025.acl-long)

Challenge:	Existing methods for logical reasoning with large language models suffer from insufficient rule semantic grounding and weak rule application mechanisms.
Approach:	They propose a theory-of-mind driven neuro-symbolic reasoning framework that integrates natural language and symbolic representations throughout the reasoning process.
Outcome:	The proposed model surpasses state-of-the-art models in reasoning accuracy and flexibility.

Copied to clipboard

Jiahao Yuan, Dehui Du, Hao Zhang, Zixiang Di, Usman Naseem

Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Models (2025.findings-acl)

Challenge:	Existing methods to improve LLMs’ logical capabilities involve traceable or verifiable logical sequences that generate more reliable responses yet increase computational costs, or introduce rigid logic template rules, reducing flexibility.
Approach:	They propose a plug-and-play reasoning framework that enhances LLMs' logical reasoning abilities during the warm-up phase prior to batch inference.
Outcome:	The proposed framework surpasses baselines in both reasoning accuracy and efficiency.

Copied to clipboard

Yingqian Cui, Pengfei He, Jingying Zeng, Hui Liu, Xianfeng Tang, Zhenwei Dai, Yan Han, Chen Luo, Jing Huang, Zhen Li, Suhang Wang, Yue Xing, Jiliang Tang, Qi He

Self-Correcting RAG: Enhancing Faithfulness via MMKP Context Selection and NLI-Guided MCTS (2026.findings-acl)

Challenge:	Chain-of-Thought (CoT) reasoning has improved the performance of large language models (LLMs) however, the detailed reasoning process in CoT often incurs long generation times and high computational costs due to the inclusion of unnecessary steps.
Approach:	They propose a method to identify critical reasoning steps using perplexity as a measure of their importance.
Outcome:	The proposed method achieves a better balance between reasoning accuracy and efficiency of CoT.

Copied to clipboard

Shijia Xu, Zhou Wu, Xiaolong Jia, Yu Wang, Kai Liu, April Xiaowen Dong

MultiLingPoT: Boosting Mathematical Reasoning in LLMs through Multilingual Program Integration (2025.findings-emnlp)

Challenge:	Existing approaches to retrieval-augmented generation still face problems with low context utilization and frequent hallucinations.
Approach:	They propose a framework that reformulates retrieval and generation as constrained optimization and path planning.
Outcome:	The proposed framework significantly improves reasoning accuracy on complex queries while reducing hallucinations.

Copied to clipboard

Nianqi Li, Zujie Liang, Siyu Yuan, Jiaqing Liang, Feng Wei, Yanghua Xiao

MPBench: A Comprehensive Multimodal Reasoning Benchmark for Process Errors Identification (2025.findings-acl)

Challenge:	Program-of-Thought is an important way for LLMs to solve mathematical problems.
Approach:	They propose a multilingual programme reasoning method that uses program instead of natural language in reasoning and proposes to integrate multilingual integration into the training and inference.
Outcome:	The proposed method improves individual language’s reasoning accuracy by 2.5% and improves performance by 8%.

Copied to clipboard

xu Zhao Pan, Pengfei Zhou, Jiaxin Ai, Wangbo Zhao, Kai Wang, Xiaojiang Peng, Wenqi Shao, Hongxun Yao, Kaipeng Zhang

Don’t Get Lost in the Trees: Streamlining LLM Reasoning by Overcoming Tree Search Exploration Pitfalls (2025.acl-long)

Challenge:	Existing benchmarks of large language models focus on error detection, neglecting other scenarios like reasoning search.
Approach:	et al. propose a multi-task, multimodal benchmark to assess effectiveness of PRMs . step correctness, answers aggregation and reasoning process search are evaluated . ethical principles of MPBench are based on a set of evaluation paradigms based in a text-based benchmark .
Outcome:	a new benchmark assesses the effectiveness of large language models (LLMs) in multiple scenarios . it uses three evaluation paradigms to assess the effectiveness and compares them with existing models . a the proposed model improves reasoning accuracy by providing stepwise feedback for multi-step reasoning results .

Copied to clipboard

Ante Wang, Linfeng Song, Ye Tian, Dian Yu, Haitao Mi, Xiangyu Duan, Zhaopeng Tu, Jinsong Su, Dong Yu

Green & Sustainable NLP

Challenge:	Recent advances in tree search algorithms guided by verifiers have significantly enhanced the reasoning capabilities of large language models (LLMs), but at the cost of increased computational resources.
Approach:	They propose an e ffici ent tree sear ch framework that is a plug-and-play system compatible with various tree search algorithms.
Outcome:	The proposed framework reduces computational costs and prioritizes resource allocation to harder tasks (Levels 3-4) over simpler ones (Level 1-2), addressing both over-exploration in basic problems and under-exploation in complex cases.

Graph Counselor: Adaptive Graph Exploration via Multi-Agent Synergy to Enhance LLM Reasoning (2025.acl-long)

Copied to clipboard

Junqi Gao, Xiang Zou, Ying Ai, Dong Li, Yichen Niu, Biqing Qi, Jianxing Liu

Diffuse Thinking: Exploring Diffusion Language Models as Efficient Thought Proposers for Reasoning (2026.acl-long)

Challenge:	Existing methods for enhancing LLM reliability suffer from inefficient information aggregation and rigid reasoning schemes.
Approach:	They propose a method that explicitly models external knowledge integration capabilities by explicitly modeling knowledge relationships.
Outcome:	The proposed method outperforms existing methods in multiple graph reasoning tasks.

Copied to clipboard

Chenyang Shao, Sijian Ren, Fengli Xu, Yong Li

Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketching (2025.emnlp-main)

Challenge:	Large language models (LLMs) have demonstrated strong capabilities in complex reasoning tasks, yet their autoregressive generation paradigm makes it computationally prohibitive to explore diverse reasoning paths.
Approach:	They propose a framework that combines diffusion-based generation with autoregressive evaluation to efficiently generate diverse intermediate reasoning thoughts and employ LLMs as evaluators to assess and select candidates based on their plausibility and correctness.
Outcome:	The proposed framework improves inference efficiency while maintaining competitive or superior reasoning accuracy.

Copied to clipboard

Simon A. Aytes, Jinheon Baek, Sung Ju Hwang

Entrospect: Information-Theoretic Self-Reflection Elicits Better Response Refinement of Small Language Models (2025.findings-acl)

Challenge:	Recent advances in large language models (LLMs) have enabled strong reasoning capabilities through Chain-of-Thought (CoT) prompting.
Approach:	They propose a framework that integrates cognitively inspired reasoning paradigms with linguistic constraints to reduce token usage while preserving reasoning accuracy.
Outcome:	The proposed framework reduces token usage while preserving reasoning accuracy across 18 reasoning datasets across multiple domains, languages, and modalities.

Copied to clipboard

Tianqiang Yan, Ziqiao Lin, Lin Zhang, Zhenglong Sun, Yuan Gao

PlanRAG-Audio: Planning and Retrieval Augmented Generation for Long-form Audio Understanding (2026.findings-acl)

Challenge:	Existing approaches to self-reflection fail to deliver robust response refinement for models with parameter sizes of 10 billion or smaller.
Approach:	They propose to redesign Self-Refine and introduce an information-theoretic framework based on Chain-of-Thought prompt engineering to improve self-reflection in Small Language Models.
Outcome:	The proposed framework improves reasoning accuracy and computational efficiency by up to 36.2% under identical model and data settings.

Copied to clipboard

Masao Someki, Chien-yu Huang, Siddhant Arora, Samuele Cornell, Markus Müller, Nathan Susanj, Rupak Vignesh Swaminathan, Grant Strimel, Jing Liu, Shinji Watanabe

Hidden States as Early Signals: Step-level Trace Evaluation and Pruning for Efficient Test-Time Scaling (2026.findings-acl)

Challenge:	Long-form audio understanding poses significant challenges due to the extreme length of audio sequences and the need to reason over heterogeneous acoustic cues distributed over time.
Approach:	They propose a retrieval-augmented generation framework for scalable long-form audio understanding . planRAG-Audio explicitly plans which modalities and temporal spans are required for a given query .
Outcome:	Experiments show that planRAG-Audio reduces the length of inputs for long-form audio models . the proposed framework can efficiently reason over long-term speech data .

Copied to clipboard

Zhixiang Liang, Beichen Huang, Zheng Wang, Minjia Zhang

Learning to Reason Over Time: Timeline Self-Reflection for Improved Temporal Reasoning in Language Models (2025.acl-long)

Challenge:	Existing approaches to speed up parallel scaling have relied on similarity-based or confidence-based pruning, but these signals do not reliably indicate trace quality.
Approach:	They propose a pruning framework that evaluates reasoning steps using hidden states and dynamically prunes unpromising traces during generation.
Outcome:	The proposed framework reduces end-to-end inference latency by 45%–70% on average compared to self-consistency while improving reasoning accuracy.

Copied to clipboard

Adrián Bazaga, Rexhina Blloshmi, Bill Byrne, Adrià de Gispert

Towards the Roots of the Negation Problem: A Multilingual NLI Dataset and Model Scaling Analysis (2025.findings-emnlp)

Challenge:	Large Language Models struggle with temporal reasoning, which requires processing time-related information such as event sequencing, durations, and inter-temporal relationships.
Approach:	They propose a framework that enhances the temporal reasoning abilities of Large Language Models (LLMs) by combining timeline construction with iterative self-reflection.
Outcome:	The proposed framework improves the temporal reasoning abilities of large language models and improves traceability of the inference process.

Copied to clipboard

Tereza Vrabcová, Marek Kadlčík, Petr Sojka, Michal Štefánik, Michal Spiegel

Robustness in NLP

Challenge:	Negations are key to determining sentence meaning, making them essential for logical reasoning.
Approach:	They construct and publish two new textual entailment datasets in four languages with paired examples differing in negation.
Outcome:	The results show that increasing the model size may improve the models’ ability to handle negations.

Evo-PI: Aligning Medical Reasoning via Evolving Principle-Guided Supervision (2026.acl-long)

Copied to clipboard

Xianda Zheng, Huan Gao, Meng-Fen Chiang, Michael J. Witbrock, Kaiqi Zhao, Shangyang Li

Exploration-Exploitation Reshaping towards Efficient Reasoning for Large Language Models (2026.findings-acl)

Challenge:	Existing models with static prompts, rules, or reward models are constrained by static supervision, which often fails to shape the underlying reasoning process, leading to brittle generalization and performance saturation in complex decision-making tasks.
Approach:	They propose a principle-centric learning framework that treats reasoning principles as explicit, language-based supervision signals that can be generated, evaluated, and iteratively evolved.
Outcome:	The proposed framework treats reasoning principles as explicit, language-based supervision signals that can be generated, evaluated, and iteratively evolved.

Copied to clipboard

Yufeng Shi, Weilin Luo, Yuxiang Zhang, Zongmeng Zhang, Haoyang Liu, Yubing Wang, Bin Wang, Wengang Zhou, Houqiang Li

DARL: Encouraging Diverse Answers for General Reasoning without Verifiers (2026.findings-acl)

Challenge:	Large Reasoning Models (LRMs) are constrained by the overthinking issue.
Approach:	They propose a policy optimization framework that reshapes the exploration and exploitation through two core components: self-imitation and self-guidance exploration.
Outcome:	The proposed model achieves superior reasoning efficiency without compromising overall accuracy.

Copied to clipboard

Chongxuan Huang, Lei Lin, Xiaodong Shi, Wenping Hu, Ruiming Tang

Unlocking the Black Box of Latent Reasoning: An Interpretability-Guided Approach to Intervention (2026.acl-long)

Challenge:	Recent efforts such as RLPR have extended RLVR to general domains, enabling training on broader datasets and achieving improvements over RL PR.
Approach:	They propose a framework that encourages the generation of diverse answers within a controlled deviation range from the reference while preserving alignment with it.
Outcome:	Extensive experiments on 13 benchmarks show that DARL surpasses RLPR in both reasoning accuracy and output diversity.

Copied to clipboard

Shuochen Chang, Tong Bai, Xiaofeng Zhang, Qianli Ma, Qingyang Liu, Zhaohe Liao, Yibo Miao, Li Niu

GAM: Hierarchical Graph-based Agentic Memory for LLM Agents (2026.acl-long)

Challenge:	Existing methods for large language models (LLMs) lack a coherent representation of reasoning steps.
Approach:	They propose a set of latent reasoning interventions that enable latent thinking and decode-time interventions that refine the latent process by imposing the identified geometric and semantic priors.
Outcome:	The proposed models unlock latent capabilities and improve reasoning accuracy without any parameter updates.

Copied to clipboard

Zhaofen Wu, Hanrong Zhang, Fulin Lin, Wujiang Xu, Xinran Xu, Yankai Chen, Henry Peng Zou, Shaowen Chen, Weizhi Zhang, Xue Liu, Philip S. Yu, Hongwei Wang

Structured Data in NLP

Challenge:	Current unified stream-based memory systems facilitate context updates but remain vulnerable to interference from transient noise.
Approach:	They propose a hierarchical Graph-based Agentic Memory framework that explicitly decouples memory encoding from consolidation to resolve conflict between rapid context perception and stable knowledge retention.
Outcome:	The proposed framework outperforms state-of-the-art benchmarks on LoCoMo and LongDialQA.

Self-Consistency from Only Two Samples: CoT–PoT Ensembling for Efficient LLM Reasoning (2026.findings-acl)

Copied to clipboard

Raman Saparkhan, Majd Hawasly, Md Rizwan Parvez, Mohammad Raza

From Detection to Understanding: Multi-Turn Reasoning for Video Misinformation Analysis (2026.acl-long)

Challenge:	Self-consistency (SC) is a popular technique for improving the reasoning accuracy of large language models but it comes at a high computational cost due to extensive sampling.
Approach:	They propose a hybrid ensembling approach that leverages the complementary strengths of Chain-of-Thought and Program-of -Thus . they propose encapsulating two different modes of reasoning to create a single output and a final answer is selected as the most frequently occurring one among these outputs.
Outcome:	The proposed approach reduces the number of samples required for SC by 9.3x . the majority of tasks can be addressed with only two samples, which has not been possible with prior methods.

Copied to clipboard

Zhi Zeng, Jiaying Wu, Minnan Luo, Di Zhang, Yifei Yang, Xiangzheng Kong, Herun Wan, Zihan Ma

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

Challenge:	Existing benchmarks focus on binary veracity judgments and do not evaluate process-level justifications for misinformation models.
Approach:	They propose a video misinformation analysis benchmark that assesses reasoning in video misinterpretation.
Outcome:	The proposed framework improves reasoning accuracy and explanation quality compared to existing models . it covers 12 fine-grained deception categories and progresses from perceptual attribution to intent and persuasion analysis.

Copied to clipboard

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

Minimal Free Resolution Guided Adaptive Tree Reasoning (2026.acl-long)

Challenge:	Large Language Models (LLMs) are reshaping recommender systems by leveraging extensive world knowledge and semantic reasoning to interpret user intent.
Approach:	They propose a single-agent Trajectory-Aligned Recommender to integrate reasoning capabilities into a model by a multi-agend teacher system.
Outcome:	The proposed model surpasses its teacher by 8.7% to 39.5% while eliminating iterative latency.

Copied to clipboard

Dezhao Tang, Meihan Liu, Yulai Tong, Guan Yuan, Qiuyan Yan