AtManRL: Towards Faithful Reasoning via Differentiable Attention Saliency

arXiv:2604.16158 · cs.CL, cs.AI, cs.LG · Submitted 2026-04-17 · Read on arXiv

cs.CL, cs.AI, cs.LG

Submitted: 2026-04-17

Updated: 2026-10-22

Terminology

• Trace — 30× in this paper · explained in 2 episodes
• Faithfulness — 14× in this paper · explained in 2 episodes
• Reasoning Traces — 9× in this paper · explained in 4 episodes
• Reinforcement Learning — 8× in this paper · explained in 4 episodes
• GRPO — 7× in this paper · explained in 6 episodes
• Accuracy — 6× in this paper · explained in 1 episode
• Adam — 5× in this paper · explained in 2 episodes
• Interpretability — 5× in this paper · explained in 15 episodes
• Chain-of-Thought (CoT) — 4× in this paper · explained in 5 episodes
• Composition — 3× in this paper · explained in 2 episodes
• Distillation — 3× in this paper · explained in 2 episodes
• Language Models (LLMs) — 3× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 3× in this paper · explained in 8 episodes
• Self-Distillation — 3× in this paper · explained in 7 episodes
• Transformer — 3× in this paper · explained in 2 episodes
• Chain-of-Thought (CoT) Reasoning — 2× in this paper · explained in 5 episodes
• Chain-of-Thought Prompting — 2× in this paper · explained in 2 episodes
• Consistency — 2× in this paper · explained in 3 episodes
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Weight Decay — 2× in this paper · explained in 2 episodes
• Adaptability — 1× in this paper · explained in 1 episode
• Chain-of-Thought (CoT) Prompting — 1× in this paper · explained in 4 episodes
• Credit assignment — 1× in this paper · explained in 3 episodes
• Embeddings — 1× in this paper · explained in 2 episodes
• Explainability — 1× in this paper · explained in 4 episodes
• GRPO Objective — 1× in this paper · explained in 2 episodes
• Latent Reasoning — 1× in this paper · explained in 3 episodes
• OPTIMIZING SALIENCY VIA REINFORCEMENT LEARNING — 1× in this paper
• Papers — 1× in this paper · explained in 3 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Reinforcement Learning (RL) — 1× in this paper · explained in 25 episodes
• Reward Shaping — 1× in this paper · explained in 2 episodes
• Reward Structure — 1× in this paper · explained in 2 episodes
• Self-Consistency — 1× in this paper · explained in 3 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Faithfulness vs. Plausibility: On the (Un)Reliability of Explanations from Large Language Models
• Training Verifiers to Solve Math Word Problems
• The Llama 3 Herd of Models
• Reinforcement Learning via Self-Distillation
• Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?
• Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?
• Measuring Faithfulness in Chain-of-Thought Reasoning
• Let's Verify Step by Step
• Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability
• Decoupled Weight Decay Regularization
• OpenAI o1 System Card
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• Ignore the KL Penalty! Boosting Exploration on Critical Tokens to Enhance RL Fine-Tuning
• Qwen3 Technical Report

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering