Papers with reasoning processes
ReasonGraph: Visualization of Reasoning Methods and Extended Inference Paths (2025.acl-demo)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) reasoning processes are complex and lack of organized visualization tools creates barriers to understanding, evaluation, and improvement. |
| Approach: | They propose a web-based platform for visualizing and analyzing LLM reasoning processes. |
| Outcome: | The proposed platform shows high parsing reliability, efficient processing, and excellent usability across various downstream applications. |
Can Large Language Models Identify Authorship? (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated exceptional capacity for reasoning and problem-solving, but their potential in authorship analysis remains under-explored. |
| Approach: | They propose to integrate explicit linguistic features into LLMs to provide explanations into their reasoning processes. |
| Outcome: | The proposed models demonstrate their ability to perform zero-shot, end-to-end authorship verification effectively and provide explainability through explicit linguistic features. |
The Price of Thought: A Multilingual Analysis of Reasoning, Performance, and Cost of Negotiation in Large Language Models (2026.findings-eacl)
Copied to clipboard
Sherzod Hakimov, Roland Bernard, Tim Leiber, Karl Osswald, Kristina Richert, Ruilin Yang, Raffaella Bernardi, David Schlangen
| Challenge: | Negotiation is a fundamental challenge for AI agents as it requires an ability to reason strategically, model opponents, and balance cooperation with competition. |
| Approach: | They propose to use a self-play setup to compare commercial and open-weight large language models to their vanilla counterparts in three different languages to examine trade-offs between performance and cost. |
| Outcome: | The proposed model improves GPT-5's performance by 31.4 % while increasing its cost by nearly 400 %. |
Prediction or Comparison: Toward Interpretable Qualitative Reasoning (2021.findings-acl)
Copied to clipboard
| Challenge: | Qualitative relationships are a significant portion of textual knowledge . current approaches use semantic parsers to transform natural language inputs into logical expressions or a "black-box" model to solve them in one step. |
| Approach: | They propose to use neural network modules to simulate qualitative reasoning tasks . they use two qualitative reasoning question answering datasets to test their methods . |
| Outcome: | Experiments on two qualitative reasoning question answering datasets show the proposed methods are general and general and interpretable. |
The Role of Exploration Modules in Small Language Models for Knowledge Graph Question Answering (2025.acl-srw)
Copied to clipboard
| Challenge: | Existing methods to integrate knowledge graphs into large language models often rely on proprietary or extremely large models . |
| Approach: | They propose to integrate knowledge graphs into reasoning processes of large language models . they propose to use simple and efficient exploration modules to handle knowledge graph traversal . |
| Outcome: | The proposed modules improve the performance of small language models on knowledge graph question answering tasks. |
KAT: A Knowledge Augmented Transformer for Vision-and-Language (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing methods for knowledge retrieval and answer prediction have left open questions about the quality and relevance of the retrieved knowledge and how the reasoning processes over implicit and explicit knowledge should be integrated. |
| Approach: | They propose a Knowledge Augmented Transformer which integrates both implicit and explicit knowledge in an encoder-decoder architecture while simultaneously reasoning over both knowledge sources during answer generation. |
| Outcome: | The proposed model achieves a strong state-of-the-art (+6% absolute) on the open-domain multimodal task of OK-VQA. |
Enhancing Large Language Models in Coding Through Multi-Perspective Self-Consistency (2024.acl-long)
Copied to clipboard
| Challenge: | Existing work utilizes verification properties to verify and re-rank solutions in a majority voting manner, but this assumption may not hold. |
| Approach: | They propose a multi-perspective self-consistency framework that incorporates both inter- and intra-consistency across outputs from multiple perspectives. |
| Outcome: | The proposed framework significantly boosts performance of foundation models on various benchmarks, including HumanEval (+15.91%), MBPP (+6.43%) and CodeContests (+9.37%). |
ReasVQA: Advancing VideoQA with Imperfect Reasoning Process (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing approaches to VideoQA often fail when complex reasoning or temporal relationships are involved. |
| Approach: | They propose a method that leverages reasoning processes generated by Multimodal Large Language Models to improve VideoQA models. |
| Outcome: | The proposed method improves VideoQA models on three benchmarks. |
LogicAsker: Evaluating and Improving the Logical Reasoning Ability of Large Language Models (2024.emnlp-main)
Copied to clipboard
Yuxuan Wan, Wenxuan Wang, Yiliu Yang, Youliang Yuan, Jen-tse Huang, Pinjia He, Wenxiang Jiao, Michael Lyu
| Challenge: | LogicAsker examines and improves the reasoning abilities of large language models such as ChatGPT and GPT-4. |
| Approach: | They propose a set of atomic reasoning skills grounded in propositional and predicate logic to examine and improve the reasoning abilities of large language models such as ChatGPT and GPT-4. |
| Outcome: | The proposed approach improves reasoning abilities in large language models such as ChatGPT and GPT-4 by up to 5%. |
Enhancing Zero-shot Chain of Thought Prompting via Uncertainty-Guided Strategy Selection (2025.coling-main)
Copied to clipboard
Shanu Kumar, Saish Mendke, Karody Lubna Abdul Rahman, Santosh Kurasa, Parag Agrawal, Sandipan Dandapat
| Challenge: | Existing methods for chain-of-thought (CoT) prompting are limited by handcrafted demonstrations and trigger phrases are prone to inaccuracies. |
| Approach: | They propose a method that generates rationales using a trigger phrase to select effective demonstrations without accessing model parameters. |
| Outcome: | The proposed method outperforms existing methods across four reasoning benchmarks and is robust and scalable. |
Aristotle: Mastering Logical Reasoning with A Logic-Complete Decompose-Search-Resolve Framework (2025.acl-long)
Copied to clipboard
Jundong Xu, Hao Fei, Meng Luo, Qian Liu, Liangming Pan, William Yang Wang, Preslav Nakov, Mong-Li Lee, Wynne Hsu
| Challenge: | Existing systems fail to fully leverage the structure of logical tasks throughout the reasoning process, causing bottlenecks in efficiency and efficacy. |
| Approach: | They propose a logic-complete reasoning framework, Aristotle, which integrates symbolic expressions and logical rules into the entire reasoning process. |
| Outcome: | The proposed framework outperforms state-of-the-art reasoning frameworks in accuracy and efficiency. |
Domaino1s: Guiding LLM Reasoning for Explainable Answers in High-Stakes Domains (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing Large Language Models (LLMs) generate brief answers without reasoning processes and explanations. |
| Approach: | They propose supervised fine-tuning and tree search to enhance LLMs’ reasoning capabilities on domain tasks. |
| Outcome: | The proposed model improves on stock investment recommendation and legal reasoning QA tasks. |
A Compliance Checking Framework Based on Retrieval Augmented Generation (2025.coling-main)
Copied to clipboard
| Challenge: | Existing text-based compliance checking methods are limited by their flexibility and lack structure. |
| Approach: | They propose a text-based compliance checking framework based on Retrieval-Augmented Generation that integrates a static layer for storing factual knowledge, a dynamic layer for retrieval and reasoning, and an eventic graph to structurally describe regulatory information. |
| Outcome: | The proposed framework consistently achieves state-of-the-art results across various scenarios surpassing baselines. |
Stepwise Reasoning Disruption Attack of LLMs (2025.acl-long)
Copied to clipboard
Jingyu Peng, Maolin Wang, Xiangyu Zhao, Kai Zhang, Wanyu Wang, Pengyue Jia, Qidong Liu, Ruocheng Guo, Qi Liu
| Challenge: | Existing attacks on LLM reasoning are constrained by specific settings or lack of imperceptibility, limiting their feasibility and generalizability. |
| Approach: | They propose a stepwise rEasoning error disruption attack that subtly injects errors into prior reasoning steps to mislead the model into producing incorrect subsequent reasoning and final answers. |
| Outcome: | The proposed attack is compatible with zero-shot and few-shot settings, maintains the natural reasoning flow, and ensures covert execution without modification of the instruction. |
SETI: Systematicity Evaluation of Textual Inference (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing pre-trained language models (PLMs) have shown remarkable performance on this task, but little is known about their ability to address compositional generalization. |
| Approach: | They propose a benchmark to evaluate pre-trained language models' systematicity in the domain of textual inference. |
| Outcome: | The proposed benchmark evaluates pre-trained language models on six widely used PLMs. |
MetaMem: Evolving Meta-Memory for Knowledge Utilization through Self-Reflective Symbolic Optimization (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing memory systems can support long-horizon human-LLM interactions by persisting historical interactions beyond limited context windows. |
| Approach: | They propose a framework that augments memory systems with a self-evolving meta-memory . meta-meso is iteratively distilling transferable knowledge utilization experiences . results show MetaMem outperforms strong baselines by over 3.6% . |
| Outcome: | The proposed framework outperforms baselines by over 3.6% in the long-horizon human-LLM interaction. |
code-transformed: The Influence of Large Language Models on Code (2026.findings-eacl)
Copied to clipboard
| Challenge: | Using Large Language Models, code generation capabilities have transformed programming practices. |
| Approach: | They analyze 20,000 GitHub repositories linked to arXiv papers published between 2020 and 2025 . they identify measurable trends in the evolution of coding style that align with LLM-generated code . |
| Outcome: | The proposed study examines 20,000 GitHub repositories linked to arXiv papers . it finds that LLMs influence code style, and that they can be observed in real-world code . |
AQuilt: Weaving Logic and Self-Inspection into Low-Cost, High-Relevance Data Synthesis for Specialist LLMs (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to synthesis large language models often suffer from performance limitations and high computational costs. |
| Approach: | They propose a framework for constructing instruction-tuning data from unlabeled data for any specialized domains from corresponding unlabed data. |
| Outcome: | The proposed framework is comparable to DeepSeek-V3 while utilizing just 17% of the production cost. |
MLeVLM: Improve Multi-level Progressive Capabilities based on Multimodal Large Language Model for Medical Visual Question Answering (2024.findings-acl)
Copied to clipboard
Dexuan Xu, Yanyuan Chen, Jieyi Wang, Yue Huang, Hanpin Wang, Zhi Jin, Hongxing Wang, Weihua Yue, Jing He, Hang Li, Yu Huang
| Challenge: | Existing MVQA models ignore multi-level progressive capabilities due to unspecific data and plain architecture. |
| Approach: | They propose a multi-level visual language model for medical visual question answering (MVQA) which covers multi- level questions and answers as well as reasoning processes from visual clues to semantic cognition. |
| Outcome: | The proposed model outperforms existing medical multimodal large language models on a multi-level instruction dataset and a feature alignment module. |
VisDoM: Multi-Document QA with Visually Rich Elements Using Multimodal Retrieval-Augmented Generation (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing benchmarks for document QA for visually rich documents outperform unimodal and long-context LLMs by 12-20%. |
| Approach: | They propose a multimodal Retrieval Augmented Generation approach that integrates visual and textual retrieval with linguistic reasoning. |
| Outcome: | The proposed approach outperforms unimodal and long-context LLM benchmarks for document QA by 12-20%. |
NOAHQA: Numerical Reasoning with Interpretable Graph Question Answering Dataset (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing question answering datasets lack numerical reasoning and reasoning processes . current research on numerical reasoning focuses on simple calculations . |
| Approach: | They propose a conversational and bilingual question answering dataset with numerical reasoning with compound mathematical expressions. |
| Outcome: | The proposed model achieves 55.5 exact match scores while human performance is 89.7. |
Did the Models Understand Documents? Benchmarking Models for Language Understanding in Document-Level Relation Extraction (2023.acl-long)
Copied to clipboard
| Challenge: | Document-level relation extraction (DocRE) models achieve consistent performance gains in DocRE, but their underlying decision rules are still understudied. |
| Approach: | They propose to use annotations to provide rationales for document-level relation extraction (DocRE) they then propose to apply a method to evaluate models' reasoning capabilities . |
| Outcome: | The proposed models exhibit different reasoning processes in contrast to humans . the proposed models render models more trustworthy and robust to be deployed in real-world scenarios. |
Three Minds, One Legend: Jailbreak Large Reasoning Model with Adaptive Stacked Ciphers (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing jailbreak methods struggle to balance effectiveness with robustness against adaptive safety mechanisms. |
| Approach: | They propose a novel approach that targets Large Reasoning Models through an adaptive encryption pipeline designed to overwhelm their reasoning capabilities. |
| Outcome: | The proposed approach achieves an attack success rate of 85.6% on OpenAI GPT-o4-mini, outperforming state-of-the-art baselines by a significant margin of 17.2%. |
GraphOTTER: Evolving LLM-based Graph Reasoning for Complex Table Question Answering (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods for complex table question answering are often implicit, feeding the entire table into prompts. |
| Approach: | They propose a GraphOTTER that explicitly establishes the reasoning process to pinpoint the correct answers. |
| Outcome: | The proposed method is able to identify the correct answers on two benchmark datasets and two LLM backbones. |
Enhancing Reasoning Abilities of Small LLMs with Cognitive Alignment (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to distill chain-of-thought (CoT) results from large language reasoning models (LRMs) to small models are ineffective and require substantial amount of annotated data. |
| Approach: | They propose a Critique-Rethink-Verify system for training small language reasoning models that can be critiquized according to the cognitive capabilities of smaller models. |
| Outcome: | The proposed system outperforms other methods on challenging reasoning benchmarks. |
Right for Right Reasons: Large Language Models for Verifiable Commonsense Knowledge Graph Question Answering (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing Knowledge Graph Question Answering (KGQA) methods focus on answering factual questions, leaving questions involving commonsense reasoning unaddressed. |
| Approach: | They propose a commonsense KGQA methodology that axiomatically surfaces commonsensical knowledge of Large Language Models and grounding every factual reasoning step on KG triples. |
| Outcome: | The proposed method outperforms existing methods and reduces instances of hallucination and reasoning errors. |
Reasoning with OmniThought: A Large CoT Dataset with Verbosity and Cognitive Difficulty Annotations (2026.acl-long)
Copied to clipboard
| Challenge: | Existing resources often fail to provide extensive reasoning problems with coherent CoT processes distilled from multiple teacher models. |
| Approach: | They propose a large-scale dataset featuring 2 million CoT processes generated by multiple powerful LRMs. |
| Outcome: | The proposed dataset features 2 million CoT processes and is validated by multiple powerful LRMs. |
R2-MultiOmnia: Leading Multilingual Multimodal Reasoning via Self-Training (2025.acl-long)
Copied to clipboard
| Challenge: | Recent studies have introduced eclectic strategies to enhance MLLMs’ reasoning capabilities, but they remain related to a single language. |
| Approach: | They propose a modular approach that instructs models to abstract key elements of the reasoning process and refine reasoning trajectories via self-correction. |
| Outcome: | The proposed approach improves multimodal reasoning, gets aligned performances among the languages approaching strong models and improves the model's performance. |
Lang2Act: Fine-Grained Visual Reasoning through Self-Emergent Linguistic Toolchains (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing frameworks depend on rigid, pre-defined external tools to extend perceptual capabilities of VLMs. |
| Approach: | They propose a framework that leverages self-emergent linguistic toolchains to enhance visual perception and reasoning. |
| Outcome: | The proposed framework improves the visual perception capabilities of large language models by incorporating external visual documents to address a given query. |
XplainLLM: A Knowledge-Augmented Dataset for Reliable Grounded Explanations in LLMs (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have achieved remarkable success in natural language tasks, yet understanding their reasoning processes remains a significant challenge. |
| Approach: | They propose a dataset that includes 24204 instances where each instance interprets the LLM’s reasoning behavior using knowledge graphs and graph attention networks (GAT). |
| Outcome: | The proposed explanation framework reduces hallucinations and improves grounded explanation generation in large language models. |
Accelerate Parallelizable Reasoning via Parallel Decoding within One Sequence (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for parallelizable reasoning tasks are inefficient, says a new study . generating lengthy reasoning sequences is computationally expensive and time-consuming, says the study authors . |
| Approach: | They propose a method that decodes multiple tokens per forward pass using a tree-like attention mask . their method achieves nearly 100% speedup in decoding while basically maintaining the answer quality . |
| Outcome: | Experimental results show that the method achieves nearly 100% speedup in decoding while maintaining the answer quality. |
How Likely Do LLMs with CoT Mimic Human Reasoning? (2025.coling-main)
Copied to clipboard
| Challenge: | Using chain-of-thought to elicit reasoning capabilities is not always effective and accurate. |
| Approach: | They compare the reasoning process of LLMs with humans to understand the causal chain . they find that LLM deviates from the ideal causal chain, resulting in spurious correlations . |
| Outcome: | The proposed method does not improve performance or accurately represent reasoning processes in LLMs. |
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
| 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. |
MAPO: Advancing Multilingual Reasoning through Multilingual-Alignment-as-Preference Optimization (2024.acl-long)
Copied to clipboard
| Challenge: | Existing models exhibit inconsistent reasoning abilities across different languages . existing models lack consistency across languages due to imbalance of training data . |
| Approach: | They propose a multilingual alignment-as-preference optimization framework to align reasoning processes in other languages with the dominant language. |
| Outcome: | The proposed framework improves multilingual reasoning across languages on three benchmarks. |
The Lessons of Developing Process Reward Models in Mathematical Reasoning (2025.findings-acl)
Copied to clipboard
Zhenru Zhang, Chujie Zheng, Yangzhen Wu, Beichen Zhang, Runji Lin, Bowen Yu, Dayiheng Liu, Jingren Zhou, Junyang Lin
| Challenge: | a recent study shows that process reward models can make mistakes, leading to wrong conclusions. |
| Approach: | They propose a consensus filtering mechanism that integrates MC estimation with LLM-as-a-judge to improve model performance and data efficiency. |
| Outcome: | The proposed model outperforms existing open-source alternatives and provides practical guidelines for future research. |
Interactive-KBQA: Multi-Turn Interactions for Knowledge Base Question Answering with Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | Knowledge base question answering (KBQA) is a challenging task, particularly in parsing intricate questions into executable logical forms. |
| Approach: | They propose a framework to generate logical forms through direct interaction with knowledge bases (KBs) by annotating a dataset with step-wise reasoning processes. |
| Outcome: | The proposed framework achieves competitive results on the WebQuestionsSP, ComplexWebQuestIONS, KQA Pro, and MetaQA datasets with a minimal number of examples (shots). Importantly, the proposed model supports manual intervention, allowing for the iterative refinement of LLM outputs. |
Enhancing Numerical Reasoning with the Guidance of Reliable Reasoning Processes (2024.acl-long)
Copied to clipboard
| Challenge: | Numerical reasoning is an essential ability for NLP systems to handle numeric information. |
| Approach: | They propose a numerical reasoning method that generates reliable reasoning processes by decomposing the answer formula and aim to train models to generate the process with synthesized data. |
| Outcome: | The proposed method improves on all five datasets with an average improvement of 1.8% compared with baselines and gpt-3.5-turbo. |
Self-Reflection Improves Safety of Large Reasoning Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing safety alignment methods are shallow and do not address deeper risks and attacks in reasoning processes. |
| Approach: | They propose a technique that introduces a special Self-Reflection token to enable LRMs to perform self-reflection during generation and recover from harmful outputs. |
| Outcome: | The proposed approach outperforms the baseline model in terms of safety and helpfulness, and significantly improves model safety without adversarial training. |
O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning (2026.findings-acl)
Copied to clipboard
Haotian Luo, Haiying He, Yibo Wang, Shiwei Liu, Wei Li, Xiaochun Cao, Dacheng Tao, Naiqiang Tan, Li Shen
| Challenge: | Recent long-thought reasoning models adopt extended reasoning processes similar to how humans ponder over complex problems. |
| Approach: | They propose a model that uses RL-style fine-tuning to reduce inference overhead while maintaining accuracy. |
| Outcome: | The proposed model reduces inference overhead while maintaining accuracy. |
Transparent and Coherent Procedural Mistake Detection (2025.emnlp-main)
Copied to clipboard
| Challenge: | Procedural mistake detection (PMD) is a problem of classifying whether a human user has successfully executed a task. |
| Approach: | They extend PMD to require generating visual self-dialog rationales to inform decisions . they leverage a natural language inference model to formulate two automated metrics for coherence of generated rationale. |
| Outcome: | The proposed model improves on a reframed task with a natural language inference model and a multi-faceted metrics visualization of common outcomes. |
From Parse-Execute to Parse-Execute-Refine: Improving Semantic Parser for Complex Question Answering over Knowledge Base (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for parsing knowledge-base questions into executable logical forms have not been successful on complex KBQA. |
| Approach: | They propose a new semantic parser called KoPL to model the reasoning processes . they propose 'parse-execute-refine' paradigm to unlock reasoning ability . |
| Outcome: | The proposed parser performs better than the state-of-the-art on complex KBQA . the proposed parsed-execute-refine paradigm can model complex reasoning steps . |
Enhanced Data Synthesis for LLM through Reasoning Structures Generated by Hierarchical GFlowNet (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to optimize instruction-response pairs lack a systematic design for the underlying reasoning structure. |
| Approach: | They propose a Reasoning Structure driven data Synthesis method that leverages a coarse-to-fine directed acyclic graph to construct reasoning structures efficiently. |
| Outcome: | The proposed method outperforms existing methods in 48.50%, 84.00%, 79.90% of the synthetic datasets trained on the proposed model. |
TwT: Thinking without Tokens by Habitual Reasoning Distillation with Multi-Teachers’ Guidance (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have made significant strides in problem-solving by incorporating reasoning processes, but this enhanced reasoning capability results in an increased number of output tokens during inference, leading to higher computational costs. |
| Approach: | They propose a method that internalizes explicit reasoning into the model’s habitual behavior through a Teacher-Guided compression strategy inspired by human cognition. |
| Outcome: | The proposed method reduces inference-time costs while maintaining high performance while preserving high quality and diversity of the distillation dataset. |
Improving Dialogue Discourse Parsing through Discourse-aware Utterance Clarification (2025.acl-long)
Copied to clipboard
| Challenge: | Extensive experiments on the STAC and Molweni datasets demonstrate that our approach effectively resolves ambiguities and significantly outperforms the state-of-the-art (SOTA) baselines. |
| Approach: | They propose a Discourse-aware Clarification Module (DCM) that generates clarifications for the parser through systematic clarification type reasoning and discourse goal reasoning. |
| Outcome: | Extensive experiments on the STAC and Molweni datasets demonstrate that the proposed module significantly outperforms the state-of-the-art (SOTA) framework. |
Veracity Bias and Beyond: Uncovering LLMs’ Hidden Beliefs in Problem-Solving Reasoning (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have been aligned to avoid harmful biases and stereotypes, but recent studies have revealed the superficial nature of this alignment. |
| Approach: | They propose to use large language models to avoid harmful biases and stereotypes by assigning personas to LLMs to observe decision discrepancies in social scenarios or asking them to associate specific attributes with social targets. |
| Outcome: | The proposed models attribute fewer correct solutions and more incorrect ones to African-American groups in math and coding, while Asian authorships are least preferred in writing evaluation. |
RankPrompt: Step-by-Step Comparisons Make Language Models Better Reasoners (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing solutions to reasoning tasks require extensive human annotations or fail in scenarios with inconsistent responses. |
| Approach: | They propose a new method that enables LLMs to self-rank their responses without additional resources. |
| Outcome: | The proposed method improves reasoning performance of ChatGPT and GPT-4 with 13% improvement over existing methods. |
EoT: Evolution of Thoughts for Complex Reasoning Tasks (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies focus on ensuring behavior fidelity, factuality or reliability in generated reasoning processes, but they neglect the simultaneous optimization of all three aspects for each thought. |
| Approach: | They propose a thought assessment method that is sensitive to knowledge and LLM behaviors . they use three scorers to evaluate each thought by considering domain context, semantic alignment, and behavior impact. |
| Outcome: | The proposed framework outperforms advanced approaches in knowledge-based complex tasks. |
Thinking Before You Speak: A Proactive Test-time Scaling Approach (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models often exhibit deficiencies with complex reasoning tasks, such as maths, due to the discrepancy between human reasoning patterns and those presented in training data. |
| Approach: | They propose to insert insights between consecutive reasoning steps to bridge this gap by generating insights between the next reasoning steps. |
| Outcome: | Experiments on mathematical datasets confirm the effectiveness of the proposed reasoning framework on complex problems. |
Reasoning-Guided Exploration for Online DPO (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent work has aimed to enhance reasoning capabilities of language models, but these methods are limited to domains with objectively verifiable answers. |
| Approach: | They propose a self-play framework to improve reasoning on general-domain data. |
| Outcome: | Experiments show that the proposed framework improves reasoning performance on general-domain data while maintaining competitive performance on verifiable academic benchmarks. |
ReasoningGuard: Safeguarding Large Reasoning Models with Inference-time Safety Aha Moments (2026.acl-long)
Copied to clipboard
| Challenge: | Existing defenses for Large Reasoning Models (LRMs) depend on costly fine-tuning and additional expert knowledge, which limits their scalability. |
| Approach: | They propose an inference-time safeguard for Large Reasoning Models that injects safety aha moments into the reasoning process to guide the model towards harmless yet helpful reasoning. |
| Outcome: | The proposed safeguard outperforms nine existing safeguards while avoiding common exaggerated safety issues. |
Machine Behavior in Relational Moral Dilemmas: Moral Rightness, Predicted Human Behavior, and Model Decisions (2026.findings-acl)
Copied to clipboard
| Challenge: | Human moral judgment is context-dependent and changes based on interpersonal relationships. |
| Approach: | They characterize LLM behavior using the Whistleblower’s Dilemma . they find moral rightness remains consistently fairness-oriented . |
| Outcome: | The model decisions mirror moral rightness judgments, rather than their behavioral predictions. |
TrustTable: A Neuro-Symbolic Auditing Framework for Faithful Table QA (2026.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs)-based TableQA models exhibit unfaithful behavior where correct answers are derived through erroneous reasoning paths. |
| Approach: | They propose a neuro-symbolic framework to audit LLM reasoning processes . it enforces factual grounding and ensures logical soundness by verifying reasoning chains . |
| Outcome: | The proposed framework outperforms LLM judges in majority voting and rejection sampling with process supervision. |
R2IF: Aligning Reasoning with Decisions via Composite Rewards for Interpretable LLM Function Calling (2026.acl-long)
Copied to clipboard
| Challenge: | Existing RL-based approaches to function calling are misaligned between reasoning processes and tool-call decisions. |
| Approach: | They propose a reasoning-aware RL framework for interpretable function calling . they integrate a composite reward integrating format/correctness constraints, CER, and SMV . |
| Outcome: | Experiments on BFCL/ACEBench show R2IF outperforms baselines by 34.62% with positive Average CoT Effectiveness. |
ReTraceQA: Evaluating Reasoning Traces of Small Language Models in Commonsense Question Answering (2026.acl-long)
Copied to clipboard
| Challenge: | Recent work in language modeling has led to effective SLMs with impressive performance levels across various benchmarks. |
| Approach: | They propose a benchmark that introduces process-level evaluation for commonsense reasoning tasks. |
| Outcome: | The proposed benchmarks show that large language models provide correct answers despite flawed reasoning processes in a substantial portion of cases. |
Too Long, Do Re-weighting for Efficient LLM Reasoning Compression (2026.acl-long)
Copied to clipboard
Zhong-Zhi Li, Xiao Liang, Zihao Tang, Lei Ji, Peijie Wang, Haotian Xu, Xing W, Haizhen Huang, Weiwei Deng, Yeyun Gong, Zhijiang Guo, Xiao Liu, Fei Yin, Cheng-Lin Liu
| Challenge: | Large Language Models (LLMs) have recently achieved remarkable progress on complex reasoning tasks by leveraging extended Chain-of-Thought (CoT) techniques. |
| Approach: | They propose a method that uses Extended Chain-of-Thought (EFT) to reduce the number of output tokens by nearly 40% while maintaining the accuracy of the reasoning. |
| Outcome: | The proposed method reduces the number of output tokens by nearly 40% while maintaining the accuracy of the reasoning. |
Embodied-Reasoner: Synergizing Visual Search, Reasoning, and Action for Embodied Interactive Tasks (2026.acl-long)
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
| 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. |
From Proof to Program: Characterizing Tool-Induced Reasoning Hallucinations in Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Tool-augmented Language Models can invoke external tools to solve problems beyond their parametric capacity. |
| Approach: | They propose a preference-optimization-based framework that realigns TaLMs to use tool outputs as assistive evidence. |
| Outcome: | The proposed framework improves accuracy and reasoning depth under tool use. |
ReCode: Reinforcing Code Generation with Reasoning-Process Rewards (2026.acl-long)
Copied to clipboard
| Challenge: | Bringing process-level supervision into RL often neglects optimizing reasoning quality. |
| Approach: | They propose a framework for RL that integrates reasoning-process rewards with strict execution outcomes and a benchmark comprising preference pairs of superior and inferior reasoning processes. |
| Outcome: | The proposed framework outperforms the base version of ReCode by 16.1% and reaches performance comparable to GPT-4-Turbo. |
A Multilingual Social Bias Benchmark Incorporating Thinking Processes (2026.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) can learn useful knowledge and harmful stereotypes, making bias evaluation essential. |
| Approach: | They propose a multilingual social bias benchmark that incorporates human-generated reasoning as part of the thinking process. |
| Outcome: | The proposed method demonstrates superior performance over LLM-generated methods . human-generated thinking yields higher-quality evaluations than template-based approaches . |