Pith. sign in

REVIEW 2 cited by

A Simple "Try Again" Can Elicit Multi-Turn LLM Reasoning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2507.14295 v2 pith:QWUJIRD2 submitted 2025-07-18 cs.LG cs.AI

classification cs.LGcs.AI
keywords multi-turnreasoningfeedbackmodelsanswerssingle-turnexistingproblem
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Multi-turn problem solving is critical yet challenging for Large Reasoning Models (LRMs) to reflect on their reasoning and revise from feedback. Existing Reinforcement Learning (RL) methods train large reasoning models on a single-turn paradigm with verifiable rewards. However, we observe that models trained with existing RL paradigms often lose their ability to solve problems across multiple turns and struggle to revise answers based on contextual feedback, leading to repetitive responses. We ask: can LRMs learn to reflect their answers in a multi-turn context? In this work, we find that training models with multi-turn RL using only unary feedback (e.g., "Let's try again") after wrong answers can improve both single-turn performance and multi-turn reasoning. We introduce Unary Feedback as Observation (UFO) for reinforcement learning, which uses minimal yet common unary user feedback during iterative problem solving. It can be easily applied to existing single-turn RL training setups. Experimental results show that RL training with UFO keeps single-turn performance and improves multi-turn reasoning accuracy by up to 14%, enabling language models to better react to feedback in multi-turn problem solving. To further minimize the number of turns needed for a correct answer while encouraging diverse reasoning when mistakes occur, we design reward structures that guide models to produce careful and deliberate answers in each turn. Code: https://github.com/lichengliu03/unary-feedback

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. DRIFT: Decoupled Rollouts and Importance-Weighted Fine-Tuning for Efficient Multi-Turn Optimization

    cs.LG 2026-05 unverdicted novelty 6.0 of 10

    DRIFT achieves multi-turn RL performance via offline importance-weighted SFT by leveraging the equivalence of KL-regularized RL to weighted supervised learning.

  2. Pushing Forward Pareto Frontiers of Proactive Agents with Behavioral Agentic Optimization

    cs.AI 2026-02 conditional novelty 5.0 of 10

    BAO, a behavior-enhanced SFT plus regularized RL pipeline, improves proactive agents' task performance while lowering user-involvement rate, beating UserRL baselines on three UserRL gym tasks.

Pith tools