Branching Policy Optimization: Sandbox-Native Language Agent Reinforcement Learning

• URL: http://arxiv.org/abs/2607.14171v1
• Date: Wed, 15 Jul 2026 09:37:36 GMT
• Title: Branching Policy Optimization: Sandbox-Native Language Agent Reinforcement Learning
• Abstract summary: Branching Policy Optimization (BPO) is a sandbox-native RL algorithm that snapshots the sandbox at high-entropy decision points along a backbone trajectory.<n>We prove this estimator is unbiased and has strictly lower variance than the trajectory-level baseline.<n>BPO improves success by 3.6--6.1 absolute points over GRPO and RLOO at matched compute, halves gradient-norm variance, and matches the best baseline using 38% fewer policy updates.
• Score: 14.640786318359709
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reinforcement learning has emerged as the dominant paradigm for training large language model (LLM) agents that interact with executable sandboxes. State-of-the-art algorithms such as PPO, RLOO, and GRPO inherit their rollout topology from RLHF: for each prompt, N independent trajectories are sampled from the initial state, and an advantage is computed by subtracting a group baseline. This design ignores a defining property of agent sandboxes. They are deterministic, snapshottable, and resumable from any intermediate state. We argue that this property enables a fundamentally different rollout topology: rather than N independent trees of depth T, one can construct a single tree of N leaves whose siblings share prefixes, and therefore share variance. We instantiate this idea as Branching Policy Optimization (BPO), a sandbox-native RL algorithm that (i) adaptively snapshots the sandbox at high-entropy decision points along a backbone trajectory, (ii) forks K alternative actions per branch point and rolls out each to termination, and (iii) computes per-step advantages from sibling returns rather than from independent prompts. We prove this estimator is unbiased and has strictly lower variance than the trajectory-level baseline, with the reduction equal to the prefix-explained portion of return variance. On WebShop, ALFWorld, and SWE-bench Verified with Qwen2.5-7B and Llama-3.1-8B backbones, BPO improves success by 3.6--6.1 absolute points over GRPO and RLOO at matched compute, halves gradient-norm variance, and matches the best baseline using 38% fewer policy updates.

Related papers

• Contrastive Branch Policy Optimization [16.16893757028268]
  Reinforcement learning with verifiable rewards (RLVR) enables language models to learn multi-turn interaction with external tools.<n>Existing methods tend to conflate two distinct problems: allocating a fixed rollout budget and translating branch outcomes into token-level credit.<n>We introduce Contrastive Branch Policy Optimization (CBPO), which disentangles these two problems and assigns a dedicated mechanism to each.
  arXiv  Detail & Related papers  (2026-08-25T09:25:55Z)
• GraphPO: Graph-based Policy Optimization for Reasoning Models [39.010538168884786]
  Reinforcement Learning with Verifiable Rewards (RLVR) has become a standard paradigm for enhancing the capability of large reasoning models.<n>Tree-based methods partly address this problem by sharing prefixes and comparing branches from the same prefix to provide fine-grained signals.<n>We propose GraphPO, a novel RL framework that represents rollouts as a directed acyclic graph, with reasoning steps as edges and semantic states summarized from the reasoning paths as nodes.
  arXiv  Detail & Related papers  (2026-06-17T11:37:54Z)
• TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning [52.67756371231985]
  Reinforcement learning with verifiable rewards (RLVR) is a promising approach for enhancing reasoning and agentic behavior in large language models.<n>We introduce Tree Rollout Allocation for Contrastive Exploration (TRACE), a unified rollout allocation framework.<n>Technically, TRACE allocates rollout budget to both prompt roots and intermediate prefixes that are most likely to yield mixed terminal rewards.
  arXiv  Detail & Related papers  (2026-06-09T17:16:03Z)
• DISA: Offline Importance Sampling for Distribution-Matching LLM-RL [56.9445657766829]
  We introduce DISA, which moves this calibration problem outside the RL loop.<n> DISA draws proposal trajectories offline, estimates the partition function via importance sampling, and freezes the resulting partition-function estimate.<n>On two open-weight backbones across six math and three code benchmarks, DISA matches or exceeds the online-coupled distribution-matching baseline Flow.
  arXiv  Detail & Related papers  (2026-05-17T07:14:44Z)
• Rethinking Importance Sampling in LLM Policy Optimization: A Cumulative Token Perspective [22.848847562976633]
  Existing methods face a fundamental bias-variance dilemma: token-level IS ratios, as adopted by PPO and GRPO, introduce bias by ignoring prefix state distribution mismatch.<n>We propose CTPO, which combines the cumulative token IS ratio with position-adaptive clipping that scales log-space clip bounds according to the natural $sqrtt$ growth of the cumulative log-ratio.
  arXiv  Detail & Related papers  (2026-05-08T06:35:02Z)
• Rethinking the Trust Region in LLM Reinforcement Learning [72.25890308541334]
  Proximal Policy Optimization (PPO) serves as the de facto standard algorithm for Large Language Models (LLMs)<n>We propose Divergence Proximal Policy Optimization (DPPO), which substitutes clipping with a more principled constraint.<n>DPPO achieves superior training and efficiency compared to existing methods, offering a more robust foundation for RL-based fine-tuning.
  arXiv  Detail & Related papers  (2026-02-04T18:59:04Z)
• Lookahead Tree-Based Rollouts for Enhanced Trajectory-Level Exploration in Reinforcement Learning with Verifiable Rewards [48.321707628011005]
  Lookahead Tree-Based Rollouts (LATR) is a novel rollout strategy designed to explicitly promote trajectory-level diversity.<n>LATR accelerates policy learning by 131% on average and improves final pass@1 performance by 4.2%.
  arXiv  Detail & Related papers  (2025-10-28T11:12:02Z)
• TreePO: Bridging the Gap of Policy Optimization and Efficacy and Inference Efficiency with Heuristic Tree-based Modeling [65.46347858249295]
  TreePO is a self-guided rollout algorithm that views sequence generation as a tree-structured searching process.<n>TreePO essentially reduces the per-update compute burden while preserving or enhancing exploration diversity.
  arXiv  Detail & Related papers  (2025-08-24T16:52:37Z)
• REBEL: Reinforcement Learning via Regressing Relative Rewards [59.68420022466047]
  We propose REBEL, a minimalist RL algorithm for the era of generative models.<n>In theory, we prove that fundamental RL algorithms like Natural Policy Gradient can be seen as variants of REBEL.<n>We find that REBEL provides a unified approach to language modeling and image generation with stronger or similar performance as PPO and DPO.
  arXiv  Detail & Related papers  (2024-04-25T17:20:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.