mlsanityData to 14 Oct 2026

CLEAR: Contrasting Textual Feedback with Experts and Amateurs for Reasoning

arXiv 2504.07116 ↗First posted 24 Mar 2025cs.CLcited by 0

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

We introduce CLEAR (Contrasting Textual Feedback with Experts and Amateurs for Reasoning), a novel approach to language model reasoning that leverages the strengths of a larger (expert) model and smaller (amateur) model. The expert and amateur models each provide feedback on a model's initial output and are contrasted with each other into refined feedback. This feedback is subsequently applied to iteratively improve CLEAR's responses. Our experiments demonstrate that CLEAR outperforms state-of-the-art methods in several challenging reasoning tasks, including story outline improvement (up to 19.6% relative increase in interestingness), constrained generation (up to 18.5% increase in coverage), mathematical reasoning (up to 6.7% improvement in accuracy) and mitigation of toxicity (decrease of up to 22% in toxicity).

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 26 of its 44 references that are in the map. First posted on arXiv 24 Mar 2025.

Builds on

26 of 44 references.

Closest ancestors · 2

Same direction, newest first

  1. 2023LLaMA: Open and Efficient Foundation Language Models27 Feb
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Models6 Oct

Other references · 24

  1. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  4. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  5. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,512
  6. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,255
  7. 2023Teaching Large Language Models to Self-Debugcited by 988
  8. 2023AGIEval: A Human-Centric Benchmark for Evaluating Foundation Modelscited by 716
  9. 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 609
  10. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 441
  11. 2023DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Modelscited by 376
  12. 2023Trusting Your Evidence: Hallucinate Less with Context-aware Decodingcited by 327
12 more
  1. 2022Generating Sequences by Learning to Self-Correctcited by 277
  2. 2021Latent Hatred: A Benchmark for Understanding Implicit Hate Speechcited by 217
  3. 2023REFINER: Reasoning Feedback on Intermediate Representationscited by 205
  4. 2022Re3: Generating Longer Stories With Recursive Reprompting and Revisioncited by 193
  5. 2023Contrastive Decoding Improves Reasoning in Large Language Modelscited by 74
  6. 2021Learning to Repair: Repairing model output errors after deployment using a dynamic memory of feedbackcited by 35
  7. 2021NL-EDIT: Correcting semantic parse errors through natural language interactioncited by 33
  8. 2023PREADD: Prefix-Adaptive Decoding for Controlled Text Generationcited by 29
  9. 2023Decomposing Complex Queries for Tip-of-the-tongue Retrievalcited by 24
  10. 2022DOC: Improving Long Story Coherence With Detailed Outline Controlcited by 11
  11. 2024THOUGHTSCULPT: Reasoning with Intermediate Revision and Searchcited by 11
  12. 2021Reinforcement Learning with Feedback from Multiple Humans with Diverse Skillscited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024CriticBench: Benchmarking LLMs for Critique-Correct Reasoning
  2. 2023MAF: Multi-Aspect Feedback for Improving Reasoning in Large Language Models
  3. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debate
  4. 2025BeyondBench: Contamination-Resistant Evaluation of Reasoning in Language Models
  5. 2021Improving Neural Model Performance through Natural Language Feedback on Their Explanations
  6. 2025A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility
  7. 2025RefineBench: Evaluating Refinement Capability of Language Models via Checklists
  8. 2022Training Language Models with Language Feedback