REVIEW 19 cited by
Teaching Large Language Models to Reason with Reinforcement Learning
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
Teaching Large Language Models to Reason with Reinforcement Learning
read the original abstract
Reinforcement Learning from Human Feedback (\textbf{RLHF}) has emerged as a dominant approach for aligning LLM outputs with human preferences. Inspired by the success of RLHF, we study the performance of multiple algorithms that learn from feedback (Expert Iteration, Proximal Policy Optimization (\textbf{PPO}), Return-Conditioned RL) on improving LLM reasoning capabilities. We investigate both sparse and dense rewards provided to the LLM both heuristically and via a learned reward model. We additionally start from multiple model sizes and initializations both with and without supervised fine-tuning (\textbf{SFT}) data. Overall, we find all algorithms perform comparably, with Expert Iteration performing best in most cases. Surprisingly, we find the sample complexity of Expert Iteration is similar to that of PPO, requiring at most on the order of $10^6$ samples to converge from a pretrained checkpoint. We investigate why this is the case, concluding that during RL training models fail to explore significantly beyond solutions already produced by SFT models. Additionally, we discuss a trade off between maj@1 and pass@96 metric performance during SFT training and how conversely RL training improves both simultaneously. We then conclude by discussing the implications of our findings for RLHF and the future role of RL in LLM fine-tuning.
Forward citations
Cited by 19 Pith papers
-
Logic-Regularized Verifier Elicits Reasoning from LLMs
LOVER creates an unsupervised logic-regularized verifier that reaches 95% of supervised verifier performance on reasoning tasks across 10 datasets.
-
CodeRL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment
CodeRL+ integrates variable-level execution trajectory inference into RLVR training to align textual code representations with execution semantics, delivering 4.6% relative pass@1 gains and generalization to code-reas...
-
Training Large Language Models to Reason in a Continuous Latent Space
Coconut lets LLMs perform reasoning directly in continuous latent space by recycling hidden states as inputs, outperforming standard chain-of-thought on search-intensive logical tasks with better accuracy-efficiency t...
-
Multi-Turn On-Policy Distillation with Prefix Replay
ReOPD offline-distills multi-turn agentic LLMs via teacher-prefix replay plus step-decay sampling, matching online OPD accuracy at ≥4× speed with zero tool calls.
-
When Reasoning Traces Become Performative: Step-Level Evidence that Chain-of-Thought Is an Imperfect Oversight Channel
CoT traces align with internal answer commitment in only 61.9% of steps on average, dominated by confabulated continuations after commitment has stabilized.
-
Epistemic Uncertainty for Test-Time Discovery
UG-TTT adds epistemic uncertainty measured by adapter disagreement as an exploration bonus in RL for LLMs, raising maximum reward and diversity on scientific discovery benchmarks.
-
Internalizing Outcome Supervision into Process Supervision: A New Paradigm for Reinforcement Learning for Reasoning
A new RL paradigm for reasoning where models generate their own internal process supervision from outcome feedback by recycling failed trajectories.
-
SeLaR: Selective Latent Reasoning in Large Language Models
SeLaR selectively applies latent soft reasoning in LLMs via entropy gating and contrastive regularization, outperforming standard CoT on five benchmarks without training.
-
LiFT: Does Instruction Fine-Tuning Improve In-Context Learning for Longitudinal Modelling by Large Language Models?
LiFT instruction fine-tunes LLMs with a temporal curriculum to improve in-context learning on longitudinal NLP tasks, yielding gains on out-of-distribution data and rare change events across multiple model sizes.
-
Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning
NPR trains LLMs to reason in parallel via self-distilled RL, delivering up to 24.5% performance gains and 4.6x speedups with 100% genuine parallel execution on reasoning benchmarks.
-
Beyond Memorization: Extending Reasoning Depth with Recurrence, Memory and Test-Time Compute Scaling
In a cellular automata rule-inference task designed to block memorization, neural models achieve high next-step accuracy but accuracy falls sharply with longer reasoning chains; depth, recurrence, memory, and test-tim...
-
RL-PLUS: Countering Capability Boundary Collapse of LLMs in Reinforcement Learning with Hybrid-policy Optimization
RL-PLUS is a hybrid RL approach for LLMs that combines internal exploitation with external data via importance sampling and exploration advantages to prevent capability boundary collapse and achieve gains on math and ...
-
LoVeC: Reinforcement Learning for Better Verbalized Confidence in Long-Form Generations
LoVeC uses RL to train LLMs to output verbalized numerical confidence scores for statements in long-form text, achieving better calibration than self-consistency baselines on QA datasets while being 20x faster.
-
Rewarding Progress: Scaling Automated Process Verifiers for LLM Reasoning
Process advantage verifiers trained to predict step-level progress under a distinct prover policy improve LLM reasoning accuracy by over 8% and sample efficiency by 5-6x over outcome reward models.
-
Training Language Models to Self-Correct via Reinforcement Learning
SCoRe uses multi-turn online RL with regularization on self-generated traces to improve LLM self-correction, achieving 15.6% and 9.1% gains on MATH and HumanEval for Gemini models.
-
NoisyCoconut: Counterfactual Consensus via Latent Space Reasoning
Injecting noise into LLM latent trajectories creates diverse reasoning paths whose agreement acts as a confidence signal for selective abstention, cutting error rates from 40-70% to under 15% on math tasks.
-
Measure Twice, Click Once: Co-evolving Proposer and Visual Critic via Reinforcement Learning for GUI Grounding
A co-evolving proposer-critic RL framework improves GUI grounding accuracy by letting the model critique its own proposals rendered on screenshots.
-
Internalizing Outcome Supervision into Process Supervision: A New Paradigm for Reinforcement Learning for Reasoning
A method internalizes outcome supervision into process supervision by extracting step-level learning signals from failed reasoning trajectories during reinforcement learning.
-
Learning to Reason at the Frontier of Learnability
A curriculum sampling questions with high variance in success rate improves reinforcement learning performance for LLM reasoning tasks.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.