6arXiv cs.LG (Machine Learning)·
TRIAGE: Role-typed credit assignment framework improves reinforcement learning for agentic tasks
TRIAGE is a new credit assignment framework for agentic reinforcement learning that augments standard GRPO by classifying action segments into semantic roles (decisive progress, useful exploration, no-progress infrastructure, regression) and applying role-conditioned process rewards. The approach addresses two structural blind spots of outcome-only credit: punishing useful exploration in failed rollouts and reinforcing redundant actions in successful ones. Evaluated on ALFWorld, Search-QA, and WebShop, TRIAGE improves success rates over GRPO and reduces environment-facing turns by 10-15%, with regression detection inside successful trajectories identified as the dominant contributor.
Related guides (4)
Related events (8)
TRACE: Tree-structured rollout budget allocation for efficient agentic RL training
TRACE (Tree Rollout Allocation for Contrastive Exploration) is a new framework for improving reinforcement learning with verifiable rewards (RLVR) in multi-turn agentic LLM settings. The method models each ReAct-style thought-action-observation turn as a distinct node, enabling budget allocation across both prompt-level and turn-level prefixes in a tree structure, rather than only at the prompt level. A shared predictor estimates conditional success probability at each anchor to guide allocation, enriching reward contrast within a fixed sampling budget. Empirically, TRACE improves Qwen3-14B multi-hop QA accuracy by 2.8 points over baselines at equal sampling cost.
APPO: Fine-grained branching and credit assignment for agentic RL in LLMs
Researchers introduce Agentic Procedural Policy Optimization (APPO), a reinforcement learning method that shifts branching and credit assignment from coarse tool-call boundaries to fine-grained decision points within generated sequences. APPO uses a Branching Score combining token uncertainty with policy-induced likelihood gains to select exploration points, plus procedure-level advantage scaling for credit distribution. Evaluated on 13 benchmarks, APPO improves strong agentic RL baselines by nearly 4 points while maintaining efficient tool use and interpretability. The work addresses a known weakness in multi-turn agentic RL: that influential decisions are distributed throughout sequences, not concentrated at tool-call boundaries.
Psy-CoT and RAPO: Psychology-grounded reasoning and role-aware RL for character-faithful role-playing agents
Researchers propose Psy-CoT, a chain-of-thought framework that decomposes role-playing reasoning into three psychology-grounded steps (Interaction Perception, Psychological Empathy, Logical Construction) to improve out-of-distribution generalization beyond surface mimicry. They also introduce Role-Aware Policy Optimization (RAPO), a reinforcement learning method that uses profile–token mutual information to weight gradients asymmetrically, addressing reward hacking where generic phrases receive the same signal as role-specific ones. Experiments on CoSER, CharacterBench, and CharacterEval show Psy-CoT outperforms existing role-playing CoT methods and RAPO consistently beats GRPO across model scales. The work addresses a known failure mode of SFT-based role-playing agents and proposes a targeted RL fix for reward model exploitation.
RREDCoT: Segment-level reward redistribution for chain-of-thought reasoning via self-approximated credit assignment
RREDCoT is a new method for redistributing rewards across segments of Chain-of-Thought traces during RL fine-tuning of reasoning language models, addressing the high-variance delayed-reward problem inherent in GRPO-style training. Rather than using computationally expensive Monte Carlo sampling for intermediate state value estimation, the method uses the model itself to approximate optimal reward redistribution without additional generation passes. The paper evaluates RREDCoT against MC sampling and several attribution baselines, analyzing segmentation strategies and state value estimation. This is relevant to the active research thread on improving RL fine-tuning stability and efficiency for reasoning models.
DelTA: Discriminative Token Credit Assignment for RLVR Training
DelTA introduces a discriminative token credit assignment method for reinforcement learning from verifiable rewards (RLVR) that addresses the problem of high-frequency formatting tokens dominating policy gradient updates. The method estimates per-token coefficients to amplify side-specific gradient directions and downweight shared or weakly discriminative ones, making the effective update direction more contrastive. On seven mathematical benchmarks, DelTA outperforms same-scale baselines by 3.26 and 2.62 average points on Qwen3-8B-Base and Qwen3-14B-Base respectively, with additional gains on code generation tasks.
LongTraceRL: Reinforcement Learning for Long-Context Reasoning via Search Agent Trajectories and Rubric Rewards
LongTraceRL is a new RL training framework for improving long-context reasoning in LLMs, addressing limitations of existing RLVR methods. It constructs challenging training data using multi-hop questions from knowledge graph random walks and tiered distractors derived from search agent trajectories (high-confusability: read but uncited; low-confusability: seen but unopened). A rubric reward provides entity-level process supervision along reasoning chains, applied only to correct responses to prevent reward hacking. Experiments across three LLMs (4B–30B parameters) on five long-context benchmarks show consistent improvements over strong baselines.
STARE: Token-level advantage reweighting to prevent entropy collapse in GRPO-style RL training
Researchers introduce STARE, a method addressing policy entropy collapse in GRPO-style reinforcement learning from verifiable rewards (RLVR) for LLM post-training. Through first-order gradient analysis, they identify a token-level credit assignment mismatch and propose selectively reweighting advantages for entropy-critical tokens using batch-internal surprisal quantiles plus a closed-loop entropy gate. Evaluated across 1.5B–32B models on short/long chain-of-thought and multi-turn tool use tasks, STARE outperforms DAPO and other baselines by 4–8% on AIME24/25 while sustaining stable training over thousands of steps.
AXPO: Agent Explorative Policy Optimization Addresses Thinking-Acting Gap in Multimodal Agentic Reasoning
This paper identifies a structural asymmetry in agentic reasoning called the 'Thinking-Acting Gap,' where tool use is attempted in only ~30% of rollouts under standard RL training (GRPO), and all-wrong tool-using subgroups suppress learning signals. The authors propose AXPO (Agent eXplorative Policy Optimization), which fixes the thinking prefix and resamples tool calls for all-wrong subgroups, combined with uncertainty-based prefix selection. Evaluated across nine multimodal benchmarks on Qwen3-VL-Thinking at multiple scales, SFT+AXPO outperforms SFT+GRPO by +1.8pp on both Pass@1 and Pass@4 at 8B, with the 8B model surpassing the 32B baseline on Pass@4 using 4× fewer parameters.