Challenge: Recent studies have shown the benefits to LLMs from fine-tuning golden-standard Chain-of-Thought rationales or using them as correct examples in few-shot prompting.
Approach: They propose a new benchmark to test the effectiveness of large language models by leveraging errors to enhance reasoning capabilities.
Outcome: The proposed methods can be used to fine-tune models in correct and incorrect domains, rather than tuning models to learn ground truth in traditional methods.

Similar Papers

Can LLMs Learn From Mistakes? An Empirical Study on Reasoning Tasks (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing work has shown that simple learning can enhance the chain-of-thought (CoT) reasoning of large language models.
Approach: They construct mistake-correction datasets to identify and correct mistakes in CoTs . they conclude that LLMs can learn from mistakes to enhance their CoT reasoning .
Outcome: The proposed datasets show that LLMs can learn from mistakes to enhance their CoT reasoning performance.
No Need for Explanations: LLMs can implicitly learn from mistakes in-context (2025.emnlp-main)

Copied to clipboard

Challenge: Existing literature assumes that correct answers to large language models must be accompanied by comprehensive rationales to be helpful.
Approach: They propose to show incorrect answers to Large Language Models (LLMs) as a popular strategy to improve their performance in reasoning-intensive tasks.
Outcome: The proposed approach outperforms chain-of-thought prompting in math reasoning tasks.
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.
I Learn Better If You Speak My Language: Understanding the Superior Performance of Fine-Tuning Large Language Models with LLM-Generated Responses (2024.emnlp-main)

Copied to clipboard

Challenge: Recent research has demonstrated that a large language model (LLM) can generate training data for another LLM, or for creating supplementary training materials, such as rationales.
Approach: They conduct an in-depth investigation to understand why fine-tuning an LLM with responses generated by a LLM often yields better results than using responses generated from humans.
Outcome: The proposed approach can be used to transfer knowledge from a larger model to a smaller one, or for creating supplementary training materials, such as rationales.
On the Impact of Fine-Tuning on Chain-of-Thought Reasoning (2025.naacl-long)

Copied to clipboard

Challenge: Large language models have emerged as powerful tools for general intelligence, showcasing advanced natural language processing capabilities.
Approach: They propose to use supervised fine-tuning and Quantized Low-Rank Adapters to improve LLMs' task-specific performance to address privacy and safety risks.
Outcome: The proposed model improves the accuracy of the chain-of-thought reasonings across four datasets and demonstrates that the faithfulness of CoT reasoning decreases.
Towards Objective Fine-tuning: How LLMs’ Prior Knowledge Causes Potential Poor Calibration? (2025.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) have enabled powerful domain-specific applications through supervised fine-tuning.
Approach: They propose a cognition-aware framework that applies targeted learning strategies according to the model’s prior knowledge to improve calibration.
Outcome: The proposed framework significantly improves calibration while maintaining performance, achieving an average 57% reduction in ECE compared to standard fine-tuning in Llama3-8B.
There’s No Such Thing as Simple Reasoning for LLMs (2025.findings-acl)

Copied to clipboard

Challenge: Existing work has focused on relatively complex “many-hop” reasoning problems.
Approach: They analyse the performance of fine-tuned LLMs on simple reasoning problems . they find the models remain highly brittle, being susceptible to seemingly innocent perturbations .
Outcome: The proposed models fail on simple reasoning problems, but are highly brittle . they are susceptible to seemingly innocent perturbations, such as adding duplicates to the set of premises and shuffling the order in which the premises are presented.
Current Advances in LLM Reasoning (2026.acl-tutorials)

Copied to clipboard

Challenge: This tutorial examines comprehensive evaluation strategies to assess the reasoning abilities of large language models (LLMs) advanced inference time methods and post-training methods that aim to make LLMs think more like humans are discussed in this tutorial.
Approach: This tutorial explores comprehensive evaluation strategies to assess the reasoning abilities of large language models (LLMs) and discusses two types of methods to improve models’ reasoning: advanced inference time methods, structured and self-improvement inference methods, and post-training methods, such as RLHF, DPO, and GRPO.
Outcome: This tutorial examines evaluation strategies to assess the reasoning abilities of large language models and discusses two types of methods to improve models’ reasoning.
Unraveling Misinformation Propagation in LLM Reasoning (2025.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated impressive capabilities in reasoning, but how they propagate within their reasoning process remains underexplored.
Approach: They propose a practical approach to mitigating misinformation propagation in LLMs by applying factual corrections early in the reasoning process and fine-tuning on synthesized data with early-stage corrections significantly improves reasoning factuality.
Outcome: The proposed model can correct misinformation when explicitly instructed, but fails to correct misinformation less than half the time even with explicit instructions.
Exposing the Achilles’ Heel: Evaluating LLMs Ability to Handle Mistakes in Mathematical Reasoning (2025.acl-long)

Copied to clipboard

Challenge: Existing evaluations focus on final accuracy, neglecting the critical aspect of reasoning capabilities.
Approach: They propose to evaluate LLMs’ abilities to detect and correct reasoning mistakes by using rule-based methods and smaller language models.
Outcome: The proposed model outperforms existing models such as GPT-4o and GPT4 in both accuracy and accuracy, but lacks data contamination and memorization concerns.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations