Papers with self-correction
DocPilot: Copilot for Automating PDF Edit Workflows in Documents (2024.acl-demos)
Copied to clipboard
| Challenge: | Document workflow copilot system that can understand user intent and execute tasks accordingly to help users streamline their workflows. |
| Approach: | They propose an AI-assisted document workflow copilot system capable of understanding user intent and executing tasks accordingly. |
| Outcome: | The proposed system can understand user intent and execute tasks accordingly to help users streamline their workflows. |
The Illusion of Insight in Reasoning Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing evidence suggests that reasoning models exhibit "Aha!" moments . however, it remains unclear whether such intrinsic shifts in reasoning strategy improve performance . |
| Approach: | They analyze reasoning traces, training checks, three reasoning domains to detect mid-reasoning shifts . they find that reasoning shifts are rare, do not become more frequent with training . |
| Outcome: | The proposed method detects mid-reasoning shifts and instrument training runs. |
AutoChecklist: Composable Pipelines for Checklist Generation and Scoring with LLM-as-a-Judge (2026.acl-demo)
Copied to clipboard
| Challenge: | AutoChecklist is an open-source library that unifies checklist-based evaluation into composable pipelines. |
| Approach: | They propose an open-source library that unifies checklist-based evaluation into composable pipelines. |
| Outcome: | The open-source library unifies checklist-based evaluation into composable pipelines. |
ProgCo: Program Helps Self-Correction of Large Language Models (2025.acl-short)
Copied to clipboard
| Challenge: | Existing LLMs fail to self-correct and generate correct feedback, leading to misleading refinement and failure of self-refinement. |
| Approach: | They propose a program-driven self-correction approach that uses program-based verification to self-refine initial responses without external feedback. |
| Outcome: | The proposed model achieves self-correction and can be further enhanced when combined with real program tools. |
Open-World Attribute Mining for E-Commerce Products with Multimodal Self-Correction Instruction Tuning (2025.acl-long)
Copied to clipboard
| Challenge: | Current AM methods focus on extracting attributes from unimodal text, underutilizing multimodal data. |
| Approach: | They propose a framework for multimodal self-correction instruction tuning to extract new attributes from images and text with Multimodal Large Language Models. |
| Outcome: | The proposed framework outperforms state-of-the-art methods on two datasets. |
Autoformalization in the Wild: Assessing LLMs on Real-World Mathematical Definitions (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated remarkable potential in assisting with mathematical reasoning on different downstream tasks. |
| Approach: | They propose two new tools for autoformalizing real-world mathematical definitions from Wikipedia and arXiv papers. |
| Outcome: | The proposed methods improve definitions by up to 16% and undefined errors by 43%. |
Self-Correcting Code Generation Using Small Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a recent study has demonstrated that self-correction is a powerful tool for code generation, but whether it is effective for smaller models remains unexplored. |
| Approach: | They propose a method that trains small language models to maintain correct outputs while progressively correcting incorrect outputs as turns proceed. |
| Outcome: | The proposed approach improves the ability of small language models for multi-turn code correction. |
Test-Time Scaling of Reasoning Models for Machine Translation (2026.eacl-long)
Copied to clipboard
| Challenge: | Using TTS, Reasoning Models (RMs) are able to perform tasks such as math and coding with limited results. |
| Approach: | They evaluate 12 Reasoning Models across a diverse suite of MT benchmarks, examining three scenarios: direct translation, forced-reasoning extrapolation, and post-editing. |
| Outcome: | The proposed approach improves translation quality on three domains, with inconsistent results for general-purpose RMs and performance plateauing. |
Benchmarking Language Model Creativity: A Case Study on Code Generation (2025.naacl-long)
Copied to clipboard
| Challenge: | Recent studies on LLM creativity evaluation focus on open-ended generation tasks . however, the degree to which LLMs possess and utilize creativity for problem-solving remains unclear . |
| Approach: | They propose a framework for quantifying LLM creativity that incorporates design ingredients . they introduce DENIAL PROMPTING which pushes LLMs to develop more creative solutions . |
| Outcome: | The proposed framework quantifies creativity in LLMs on Codeforces problems . it also finds that even the most creative model fails to demonstrate human-like creativity . |
VisCoder: Fine-Tuning LLMs for Executable Python Visualization Code Generation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing instruction-tuning datasets lack execution-grounded supervision and offer limited support for iterative code correction. |
| Approach: | They propose a large-scale instruction tuning dataset for Python-based visualization and self-correction. |
| Outcome: | The proposed dataset outperforms strong open-source baselines and proprietary models like GPT-4o-mini. |
Chain-of-Interactions: Multi-step Iterative ICL Framework for Abstractive Task-Oriented Dialogue Summarization of Conversational AI Interactions (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have introduced paradigm-shifting approaches in natural language processing, yet their transformative in-context learning (ICL) capabilities remain underutilized, especially in customer service dialogue summarization. |
| Approach: | They propose a single-instance, multi-step framework that orchestrates information extraction, self-correction, and evaluation through sequential interactive generation chains. |
| Outcome: | The proposed framework outperforms existing models and prompts in the customer service dialogue summarization domain. |
Confidence v.s. Critique: A Decomposition of Self-Correction Capability for LLMs (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to improve self-correction performance of Large Language Models are based on intrinsic selfcorrectione, which allows the model to check and revise its selfgenerated answers without external feedback. |
| Approach: | They propose to decompose the self-correction capability into confidence and critique capabilities and a metric for overall self-corretion capability evaluation. |
| Outcome: | The proposed method outperforms vanilla SFT and achieves much higher accuracy after self-correction. |
Self-Correction is More than Refinement: A Learning Framework for Visual and Language Reasoning Tasks (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have shown remarkable abilities, but they invariably generate flawed responses. |
| Approach: | They propose a self-correction approach that instructs VLMs to refine their outputs by allowing them to learn from their self-generated self-reference data without external feedback. |
| Outcome: | The proposed approach enables VLMs to learn from their self-generated self-correction data without relying on external feedback, facilitating self-improvement. |
Beyond Lines and Circles: Unveiling the Geometric Reasoning Gap in Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent advances in Large Language Models (LLMs) demonstrate increasing proficiency in complex mathematical and algorithmic tasks, yet their geometric reasoning skills are underexplored. |
| Approach: | They propose a framework that enhances LLMs’ reasoning potential through a multi-agent system conducting internal dialogue. |
| Outcome: | The proposed framework enhances LLMs’ reasoning potential through a multi-agent system conducting internal dialogue. |
Markov Chain of Thought for Efficient Mathematical Reasoning (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing studies have sought to enhance the mathematical reasoning capabilities of large language models. |
| Approach: | They propose a Markov Chain of Thought (MCoT) that compresses previous reasoning steps into a simplified question. |
| Outcome: | The proposed method improves efficiency and maintains comparable accuracy. |
Discourse Heuristics For Paradoxically Moral Self-Correction (2025.findings-emnlp)
Copied to clipboard
| Challenge: | moral self-correction is a promising approach for aligning output of Large Language Models with human moral values . authors show that moral self correction relies on discourse constructions that reflect heuristic shortcuts . |
| Approach: | a new method is proposed to strengthen moral self-correction using heuristics extracted from curated datasets. |
| Outcome: | a new method to strengthen moral self-correction is proposed . the proposed method is based on heuristics extracted from curated datasets. |
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. |
How to Learn in a Noisy World? Self-Correcting the Real-World Data Noise in Machine Translation (2025.findings-naacl)
Copied to clipboard
| Challenge: | Semantic misalignment, as the primary source of the noise, poses a challenge for training machine translation systems. |
| Approach: | They propose a process for simulating misalignment controlled by semantic similarity which closely resembles misaligned sentences in real-world web-crawled corpora. |
| Outcome: | The proposed model significantly improves translation performance in the presence of misalignment noise and when applied to real-world, noisy web-mined datasets, across a range of translation tasks. |
ToolRM: Towards Agentic Tool-Use Reward Modeling (2026.findings-acl)
Copied to clipboard
Renhao Li, Jianhong Tu, Yang Su, Yantao Liu, Fei Huang, Hamid Alinejad-Rokny, Derek F. Wong, Junyang Lin, Min Yang
| Challenge: | lack of reliable reward models for tool-use tasks has limited progress toward agentic AI . recent advances in agentic artificial intelligence are driven by tool-using capabilities of large language models. |
| Approach: | They propose a pipeline that constructs pairwise preference data using rule-based scoring and multidimensional sampling to build lightweight reward models. |
| Outcome: | The proposed model outperforms existing models on tool calling tasks with higher accuracy. |
Self-Reflective Generation at Test Time (2026.acl-long)
Copied to clipboard
| Challenge: | Existing self-reflection mechanisms are reactive and inefficient for large language models . a fundamental tension persists between the ability to execute complex multi-step reasoning and the ability of the model to generate coherent traces. |
| Approach: | They propose a test-time framework that reflects before generating at uncertain points . SRGen utilizes dynamic entropy thresholding to identify high-uncertainty tokens . |
| Outcome: | The proposed framework can significantly strengthen large language models' reasoning process. |
Decomposition for Enhancing Attention: Improving LLM-based Text-to-SQL through Workflow Paradigm (2024.findings-acl)
Copied to clipboard
Yuanzhen Xie, Xinzhou Jin, Tao Xie, Matrixmxlin Matrixmxlin, Liang Chen, Chenyun Yu, Cheng Lei, Chengxiang Zhuo, Bo Hu, Zang Li
| Challenge: | In-context learning of large-language models has achieved remarkable success in the field of natural language processing . however, the single-step chain-of-thought prompting approach faces challenges such as attention diffusion and inadequate performance in complex tasks like text-to-SQL. |
| Approach: | They propose a workflow paradigm method to enhance the attention and problem-solving scope of large-language models through decomposition. |
| Outcome: | The proposed method outperforms existing methods on three datasets and improves the upper limit of LLM-based approaches. |
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. |
A Probabilistic Inference Scaling Theory for LLM Self-Correction (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated the capability to refine their generated answers through self-correction, enabling continuous performance improvement over multiple rounds. |
| Approach: | They propose a probabilistic theory to model the dynamics of accuracy change and explain performance improvements observed in multi-round self-correction. |
| Outcome: | The proposed model can predict accuracy curves and improve accuracy over multiple rounds. |
Counterfactual Debating with Preset Stances for Hallucination Elimination of LLMs (2025.coling-main)
Copied to clipboard
| Challenge: | Existing solutions to alleviate hallucination have considered utilizing LLMs’ inherent reasoning abilities to alleviating hallucinism, such as self-correction and diverse sampling methods. |
| Approach: | They propose a counterfactual multi-agent debate framework that predetermines LLMs' stances to override their inherent biases for answer inspection. |
| Outcome: | Extensive experiments on four datasets of three tasks demonstrate the superiority of the proposed framework over existing methods. |
Divide-Verify-Refine: Can LLMs Self-align with Complex Instructions? (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing research shows LLMs struggle with complex instructions involving multiple constraints. |
| Approach: | They propose a framework to divide complex instructions into single constraints and prepare appropriate tools to verify responses. |
| Outcome: | The proposed framework doubles Llama3.1-8B’s constraint adherence and triples Mistral-7B’ s performance. |
Meeseeks: A Feedback-Driven, Iterative Self-Correction Benchmark evaluating LLMs’ Instruction Following Capability (2026.findings-acl)
Copied to clipboard
Jiaming Wang, Yunke Zhao, Peng Ding, Jun Kuang, Yibin Shen, Zhe Tang, Yilin Jin, ZongYu Wang, Xiaoyu Li, Xuezhi Cao
| Challenge: | Existing models lack the ability to adhere to instructions, resulting in suboptimal performance. |
| Approach: | They propose an automated iterative instruction-following benchmark with integrated feedback mechanism. |
| Outcome: | The proposed benchmark identifies erroneous components in model responses and provides feedback accurately. |
LLMs cannot find reasoning errors, but can correct them given the error location (2024.findings-acl)
Copied to clipboard
| Challenge: | Recent attempts to self-correct logical or reasoning errors often cause correct answers to become incorrect, resulting in poor performance overall. |
| Approach: | They propose to use a backtracking setup to test the correction abilities of LLMs on their mistake-finding ability to find logical mistakes. |
| Outcome: | The proposed model improves on 5 reasoning tasks, showing that it can correct logical mistakes without ground truth labels or training data. |
Agent-RewardBench: Towards a Unified Benchmark for Reward Modeling across Perception, Planning, and Safety in Real-World Multimodal Agents (2025.acl-long)
Copied to clipboard
| 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 . |
Effective Synthetic Data and Test-Time Adaptation for OCR Correction (2024.emnlp-main)
Copied to clipboard
| Challenge: | Recent research has framed the post-OCR task as a Seq2Seq Neural Machine Translation (NMT) task. |
| Approach: | They propose a method for constructing post-OCR synthetic data with different noise levels using weak supervision. |
| Outcome: | The proposed method reduces CER by 68.67% without relying on manual annotations. |
Intrinsic Self-correction for Enhanced Morality: An Analysis of Internal Mechanisms and the Superficial Hypothesis (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on the effectiveness of moral self-correction in large language models have not been conducted. |
| Approach: | They propose that moral self-correction is a computationally efficient method for reducing harmful content in LLMs. |
| Outcome: | The proposed method reduces harmful content in LLMs, but it remains under-explored . it can help LLM find shortcut to more morally correct output, the authors argue . |
Small Language Models Need Strong Verifiers to Self-Correct Reasoning (2024.findings-acl)
Copied to clipboard
Yunxiang Zhang, Muhammad Khalifa, Lajanugen Logeswaran, Jaekyeom Kim, Moontae Lee, Honglak Lee, Lu Wang
| Challenge: | Existing studies show that large language models can self-correct their outputs by generating a critique and revising it based on the critique. |
| Approach: | They propose a pipeline that prompts small language models to collect self-correction data that supports the training of self-refinement abilities. |
| Outcome: | The proposed pipeline improves the self-correction abilities of two models on five datasets spanning math and commonsense reasoning. |
Beyond Ranking: Fine-Grained Diagnostics and Self-Improvement for MLLMs (2026.acl-long)
Copied to clipboard
| Challenge: | Current paradigms rely on holistic scoring and static leaderboards to disentangle fine-grained competencies. |
| Approach: | They propose a framework to shift the focus from ranking to fine-grained diagnosis. |
| Outcome: | The proposed framework surpasses the strongest baseline by 7.92%. |
DSVD: Dynamic Self-Verify Decoding for Faithful Generation in Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to reliability of large language models often lack self-correction or use costly post-hoc verification. |
| Approach: | They propose a decoding framework that enhances generation reliability through real-time hallucination detection and efficient error correction. |
| Outcome: | Extensive experiments across five benchmarks show the proposed framework improves truthfulness and factual accuracy. |
S2R: Teaching LLMs to Self-verify and Self-correct via Reinforcement Learning (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to incentivize LLMs’ deep thinking abilities require large-scale data or significant training efforts. |
| Approach: | They introduce an efficient framework that enhances LLM reasoning by teaching models to self-verify and self-correct during inference. |
| Outcome: | The proposed framework outperforms models trained on long-CoT distilled data with 3.1k initialization samples and achieves an accuracy improvement of 51.0% to 81.6%. |
Metacognitive Self-Correction for Multi-Agent System via Prototype-Guided Next-Execution Reconstruction (2026.findings-acl)
Copied to clipboard
Xu Shen, Qi Zhang, Song Wang, Zhen Tan, Xinyu Zhao, Laura Yao, Vaishnav Tadiparthi, Hossein Nourkhiz Mahjoub, Ehsan Moradi Pari, Kwonjoon Lee, Tianlong Chen
| Challenge: | Large Language Model based multi-agent systems (MAS) excel at collaborative problem solving but remain brittle to cascading errors. |
| Approach: | They propose a metacognitive framework that enables step-level error detection and self-correction in Large Language Model based multi-agent systems (MAS) . |
| Outcome: | The proposed framework outperforms baselines on the Who When benchmark and delivers consistent gains on AgentErrorBench. |
Draft, Verify, Restore: Self-Refining Historical Inscription Restoration with a Unified MLLM (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for end-to-end historical inscription restoration rely on task-separated pipelines with irreversible error accumulation and patch-based generation that sacrifices page-level consistency. |
| Approach: | They propose a unified MLLM for end-to-end historical inscription restoration that integrates draft-guided localization and Hierarchical self-refinement to enable accurate damage localization. |
| Outcome: | The proposed model achieves superior performance in both text restoration accuracy and appearance restoration quality. |
GRAT: Guiding Retrieval-Augmented Reasoning through Process Rewards Tree Search (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods to enhance large models for multi-hop question-answering lack the ability for multipath exploration, strategic look-ahead, stepwise evaluation, and global selection. |
| Approach: | They propose an algorithm guided by Monte Carlo Tree Search and process rewards that assigns fine-grained rewards to each step in the search path. |
| Outcome: | The proposed algorithm outperforms various RAG-based methods on four multihop QA datasets and shows that it can self-train and self-update. |
Mobile-R1: Towards Interactive Capability for VLM-Based Mobile Agent via Systematic Training (2026.acl-long)
Copied to clipboard
Jihao Gu, Qihang Ai, Yingyao Wang, Pi Bu, Jingxuan Xing, Yue Cao, Zekun Zhu, Wei Jiang, Ziming Wang, Yingxiu Zhao, Ming-Liang Zhang, Jun Song, Yuning Jiang, Bo Zheng
| Challenge: | Existing approaches to training agents for visual-language models trap them in local optima, hindering exploration and error correction with the environment. |
| Approach: | They propose a hierarchical training recipe that bridges atomic action execution and strategic task completion. |
| Outcome: | The proposed training recipe bridges atomic action execution and strategic task completion. |
The Validation Gap: A Mechanistic Analysis of How Language Models Compute Arithmetic but Fail to Validate It (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have explored methods to enhance self-correction in large language models, but little attention has been given to understanding the models’ internal mechanisms underlying error detection. |
| Approach: | They propose to use a large language model to analyze arithmetic errors in four smaller-sized LLMs and identify their internal mechanisms. |
| Outcome: | The proposed models heavily rely on consistency headstextemdashattention heads that assess surface-level alignment of numerical values in arithmetic solutions. |
LongCLI-Bench: A Preliminary Benchmark and Study for Long-horizon Agentic Programming in Command-Line Interfaces (2026.findings-acl)
Copied to clipboard
Yukang Feng, Jianwen Sun, Zelai Yang, Jiaxin Ai, Chuanhao Li, Zizhen Li, Fanrui Zhang, Kang He, Rui Ma, Jifan Lin, Jie Sun, Yang Xiao, Sizhuo Zhou, Wenxiao Wu, Yiming Liu, Pengfei Liu, Shenglin Zhang, Kaipeng Zhang
| Challenge: | Existing benchmarks for agentic programming in long-horizon command-line interface tasks are limited by short task horizons, data contamination from GitHub scraping, and a lack of fine-grained evaluation metrics. |
| Approach: | They propose a benchmark to evaluate agentic capabilities across long-horizon command-line interface tasks. |
| Outcome: | The proposed benchmarks cover four engineering categories: from scratch, feature addition, bug fixing, and refactoring. |
ReflexiCoder: Teaching Large Language Models to Self-Reflect on Generated Code and Self-Correct It via Reinforcement Learning (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing iterative refinement strategies that generate solutions in a single forward pass often hit a performance ceiling on complex algorithmic tasks. |
| Approach: | They propose a reinforcement learning framework that internalizes the structured reasoning trajectory directly into the model’s weights. |
| Outcome: | The proposed framework achieves 94.51% (87.20%) on HumanEval, 81.80% (78.57%) on MBPP, 35.00% on BigCodeBench, 52.21% on LiveCodeBech, and 37.34% on CodeForces in a single-attempt setting. |
Beyond ’Aha!’: Toward Systematic Meta-Abilities Alignment in Large Reasoning Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Prior work has shown that outcome-based reinforcement learning (RL) can incidentally elicit advanced reasoning behaviors such as self-correction, backtracking, and verification. |
| Approach: | They explicitly align large reasoning models with three meta-abilities: deduction, induction, and abduction, using automatically generated, self-verifiable tasks. |
| Outcome: | The proposed model aligns models with deduction, induction, and abduction meta-abilities using automatically generated, self-verifiable tasks. |
Efficient Hallucination Detection in Automatic Code Generation (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models produce source code that appears correct and well-formed, but includes hallucinated elements that cause downstream test failures. |
| Approach: | They develop a transformer-based detector that uses LLM internal representations to identify hallucinations. |
| Outcome: | The proposed detector outperforms existing methods and unsupervised methods in the code generation domain. |