# TIC-GRPO: Provable and Efficient Optimization for Reinforcement Learning from Human Feedback

arXiv: https://arxiv.org/abs/2508.02833 · HTML: https://mlsanity.com/paper/2508.02833/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 26 of its 45 references that are in the map.
- First posted on arXiv 4 Aug 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)
- LLM alignment / RLHF / preference optimisation / [Process reward models for reasoning](https://mlsanity.com/alignment/process-reward-models-for-reasoning/)

## Builds on (most cited first, 26 of 26)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [Understanding R1-Zero-Like Training: A Critical Perspective](https://mlsanity.com/paper/2503.20783/) (2025)
- [Group Sequence Policy Optimization](https://mlsanity.com/paper/2507.18071/) (2025)
- [Alignment faking in large language models](https://mlsanity.com/paper/2412.14093/) (2024)
- [Principled Reinforcement Learning with Human Feedback from Pairwise or K-wise Comparisons](https://mlsanity.com/paper/2301.11270/) (2023)
- [FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving](https://mlsanity.com/paper/2505.17685/) (2025)
- [Mastering Complex Control in MOBA Games with Deep Reinforcement Learning](https://mlsanity.com/paper/1912.09729/) (2019)
- [GPG: A Simple and Strong Reinforcement Learning Baseline for Model Reasoning](https://mlsanity.com/paper/2504.02546/) (2025)
- [Global Convergence of Policy Gradient Methods to (Almost) Locally Optimal Policies](https://mlsanity.com/paper/1906.08383/) (2019)
- [S-GRPO: Early Exit via Reinforcement Learning in Reasoning Models](https://mlsanity.com/paper/2505.07686/) (2025)
- [A general sample complexity analysis of vanilla policy gradient](https://mlsanity.com/paper/2107.11433/) (2021)
- [Closing the Gap Between the Upper Bound and the Lower Bound of Adam's Iteration Complexity](https://mlsanity.com/paper/2310.17998/) (2023)
- [On the Global Optimum Convergence of Momentum-based Policy Gradient](https://mlsanity.com/paper/2110.10116/) (2021)
- [Adaptive Batch-Wise Sample Scheduling for Direct Preference Optimization](https://mlsanity.com/paper/2506.17252/) (2025)
- [Non-asymptotic Convergence of Adam-type Reinforcement Learning Algorithms under Markovian Sampling](https://mlsanity.com/paper/2002.06286/) (2020)
- [On the Sample Complexity and Metastability of Heavy-tailed Policy Search in Continuous Control](https://mlsanity.com/paper/2106.08414/) (2021)
- [A Comprehensive Framework for Analyzing the Convergence of Adam: Bridging the Gap with SGD](https://mlsanity.com/paper/2410.04458/) (2024)
- [Model-Based Reparameterization Policy Gradient Methods: Theory and Practical Algorithms](https://mlsanity.com/paper/2310.19927/) (2023)

## Built on by (most cited first, 11 of 11)

- [WS-GRPO: Weakly-Supervised Group-Relative Policy Optimization for Rollout-Efficient Reasoning](https://mlsanity.com/paper/2602.17025/) (2026)
- [Clip Your Sequences Fairly: Enforcing Length Fairness for Sequence-Level RL](https://mlsanity.com/paper/2509.09177/) (2025)
- [When Models Learn to Ask Why: Adaptive Causal Reasoning for Trustworthy Medical Vision-Language Models](https://mlsanity.com/paper/2603.23085/) (2026)
- [HIPO: Instruction Hierarchy via Constrained Reinforcement Learning](https://mlsanity.com/paper/2603.16152/) (2026)
- [SQL-ASTRA: Alleviating Sparse Feedback in Agentic SQL via Column-Set Matching and Trajectory Aggregation](https://mlsanity.com/paper/2603.16161/) (2026)
- [BASIS: Batchwise Advantage Estimation from Single-Rollout Information Sharing for LLM Reasoning](https://mlsanity.com/paper/2605.27293/) (2026)
- [FGRPO: Federated GRPO with Adaptive Aggregation on Non-IID Data](https://mlsanity.com/paper/2606.03094/) (2026)
- [In2x at WMT25 Translation Task](https://mlsanity.com/paper/2508.14472/) (2025)
- [LC-ERD: Mining Latent Logic for Self-Evolving Reasoning via Consistency-Regulated Reward Decomposition](https://mlsanity.com/paper/2605.24005/) (2026)
- [Spurious Advantage Hidden in GRPO](https://mlsanity.com/paper/2609.04063/) (2026)
- [Asynchronous LLM Post-Training: Group-Mass Capping and Convergence Analysis](https://mlsanity.com/paper/2610.01896/) (2026)

## Abstract (arXiv)

Group Relative Policy Optimization (GRPO) trains language models without a value critic using rewards centered within sampled response groups. We study how importance weighting, clipping, and length normalization affect its stochastic updates and propose Trajectory-level Importance-Corrected GRPO (TIC-GRPO), combining an upper importance-ratio cap with a full-trajectory likelihood ratio. Under common maximum-length normalization, a trajectory change of measure gives a score second moment proportional to the response horizon. We propagate this estimate through the actual empirical baseline and all inner updates. With separately optimized constant steps, the sampling term in TIC's stationarity bound has a linear horizon coefficient, versus order three halves for token-level GRPO2. Matching upper and lower bounds separate the capped methods' worst-case response-level variances by a factor T. The bounds track prompt count, response-group size, vocabulary size, reward scale, and score regularity; per-response-normalized GRPO also retains a length-covariance term. Experiments at two Qwen3 scales on four reasoning and coding benchmarks evaluate both mechanisms in sampled-token-normalized training.