Token-Level Credit Assignment

Overview

Token-level credit assignment studies how to turn sparse sequence-level feedback into learning signals for individual tokens, steps, or actions in language-model training. Methods include contrasting successful and failed rollouts, using uncertainty or hidden-state and attention signals, deriving token rewards from verifiers or execution traces, and coordinating token scores with critics or self-distillation. These approaches are applied to verifiable reasoning, code generation, multimodal grounding, and multi-turn tool use, where the important contribution may be concentrated in a small part of a response or trajectory. A central challenge is ensuring that dense signals genuinely predict task success: noisy teacher likelihoods, excessive reweighting, and replay variation can misdirect updates or destabilize training.

Latest papers

60 papers6 papers in the last four weeks, down 50% on the four weeks before. 0.0% of all new papers.

Nov 14, 2026

SARGO: Agreement-Gated Reinforcement and Suppression via On-Policy Self-Guidance for Visual Hallucination Mitigation

Vision-language models remain susceptible to visual hallucinations, often mixing grounded information with unsupported claims in the same response. Response-level reinforcement learning assigns a shared advantage to all tokens in a response, causing negative updates to suppress correct information alongside unsupported claims. To reduce hallucination while preserving useful visual information, we introduce SARGO (Self-guided Advantage-Routed Grounding Optimization), a post-training framework for selective reinforcement and suppression through on-policy self-guidance. Direct and reciprocal weighting gives better-supported tokens stronger reinforcement on positive updates and weaker penalties on negative updates. Agreement between overall and hallucination-specific advantages selects when to apply this guidance. At the response level, a reference-calibrated reward couples factuality with the retention and expansion of grounded information, discouraging lower hallucination scores obtained through omission. Experiments on Qwen3-VL-4B, Qwen3-VL-8B, and Gemma3-4B demonstrate that SARGO outperforms the evaluated mitigation baselines across Obj-Hal, AMBER, THRONE, and SHR evaluations. SARGO achieves these gains while preserving useful visual information and general multimodal capabilities.
Nov 9, 2026

AgentOPSD: Turn-Level Credit Assignment via Recursive Belief Revision

Reinforcement learning with verifiable rewards ordinarily spreads a single advantage over a whole trajectory, a granularity too coarse to isolate the few decisions on which success in a long, multi-turn interaction actually turns. Privileged self-distillation supplies denser supervision, but it leaves unanswered how a set of local teacher-student discrepancies should be converted into credit along a sequence. We introduce AgentOPSD, a critic-free scheme that assigns credit recursively, one interaction turn at a time. Inside a turn, the token-level log-probability gaps between a skill-conditioned self-teacher and the student are summed, and that sum acts as evidence in a recursive Bayesian update carried out in log-odds space; the resulting revisions reallocate the outcome-derived advantage across turns through a multiplier that is bounded and preserves its sign. With Qwen2.5-3B and Qwen2.5-7B as policy backbones on ALFWorld, the method improves consistently on GRPO and on self-distillation baselines, and attains 89.1% success at the 7B scale. Ablations attribute the gain to aggregating evidence at turn boundaries and to making the belief revision depend on history, rather than to the local gap by itself.
Nov 8, 2026

When Credit Turns Against Tools: Transition-Aware Credit Assignment in Agentic Reinforcement Learning

Agentic reinforcement learning with verifiable rewards trains language models through multi-turn tool interactions, yet their reasoning performance and tool-calling ability can decline during training. We trace this degradation to three coupled mechanisms: (1) tool-token imbalance produces negative overall tool credit when incorrect trajectories contain more tool tokens per turn than correct ones; (2) aggregation amplification can further strengthen this negative signal as token-mean aggregation gives longer incorrect trajectories greater weight in policy updates; and (3) failure-driven retries prolong trajectories as models tend to call tools again after execution errors, extending potential negative credit to recovery attempts. To counter this suppression, we propose Transition-Aware Credit Assignment (TACA), which targets the transition from reasoning to tool use through supervision of both call initiation and generation. Specifically, TACA uses Advantage Projection to enforce a nonnegative mean advantage on tool-call tokens within each prompt group through a minimal uniform shift. Additionally, TACA guides tool-use decisions with Decision Contrast, comparing rewards for tool-assisted and direct answers to the same prompt. Through Supervised Replay, it reinforces tool-call generation using calls from previously sampled correct trajectories. Across models and tasks, TACA improves performance and sustains more reliable tool interaction during training, raising mean accuracy over MT-GRPO on six reasoning benchmarks by 4.61 percentage points for Qwen2.5-1.5B and 3.61 for Qwen2.5-3B. Our code is available at https://github.com/tmlr-group/TACA.
Nov 4, 2026

Using Experience Twice: Dynamic Process Rubrics for Privileged Token-Level Credit Assignment

Agent trajectories often contain both successful decisions and failed attempts, yet optimization with sparse trajectory-level rewards struggles to identify which critical decisions should be reinforced or suppressed. We observe that meaningful distinctions among on-policy trajectories can be distilled into natural-language process rubrics and used as process supervision signals. Based on this insight, we propose Online Rubric Self-Distillation (ORSD), which leverages self-derived process rubrics for dense token-level supervision, thereby augmenting outcome-centric reinforcement learning. Specifically, the current policy independently generates multiple rubric bundles at each step according to the same group of trajectories, extracting candidate process rubrics from cross-trajectory behavioral differences. A frozen verifier then evaluates all rubric items against each trajectory in the group, and selects one candidate bundle based on bundle-level alignment between rubric satisfaction and task success. ORSD treats the selected rubric bundle as privileged information during training, then compares the log-likelihoods of the original tokens with and without the rubric. The resulting likelihood gap provides token-specific process credit, which is added to the original trajectory-level GRPO advantage through additive advantage shaping. ORSD thus bridges trajectory-level outcome supervision and fine-grained process supervision, enabling token-level credit assignment using process signals. Extensive experiments on ALFWorld and WebShop show that ORSD accelerates policy improvement and boosts final task performance for both Qwen3-1.7B and Qwen3-4B under the same environment-interaction budget. Our code is available at https://github.com/AnnaXChun/ORSD.
Oct 29, 2026

Distributional Reward Calibration and Budget-Neutral Credit Assignment for Multimodal RLVR

We revisit advantage estimation in Group Relative Policy Optimization (GRPO) and observe that any token-level advantage admits a two-stage factorization: a sample-level baselining step that converts scalar verifier rewards into a comparable sequence-level signal, and a token-level credit assignment step that distributes that signal over the trajectory. Existing estimators typically address only one of these stages: affine baselines preserve the shape of each task's reward law, so heavy tails and binary cliffs propagate into the gradient, while token-level credit rules tie the size of each token's update to raw confidence scales and need not conserve the total credit issued by the verifier. We propose TACE (Two-stage Advantage and Credit Estimation), a two-stage estimator that addresses both stages. At stage~(S1), we replace linear standardization with a one-dimensional optimal-transport map to a standard normal, which gives a common population reward law across tasks and caps finite-sample outliers deterministically. At stage~(S2), we rank-calibrate a stop-gradient evidence contrast within each trajectory and renormalize the clipped token weights to have mean one, so credit is redistributed across tokens while its total mass and sign are conserved. We prove population-level inter-task moment equity, a deterministic tail bound, and mass-conservation, boundedness, and sign-anchoring properties that affine baselines and unnormalized token weights do not satisfy. On a multimodal reasoning suite, TACE obtains the strongest results on a majority of the benchmarks but does not dominate every setting.
Oct 23, 2026

Direction-Aware Token Credit Assignment with Reward-Model Gradients

GRPO shares a response-level advantage across tokens, leaving unresolved how strongly each token should contribute to learning. Updates induced by the shared advantage reinforce or suppress the sampled token at each position, but these updates do not always point toward increasing reward at the token level. This motivates a work-based view of credit: just as work depends on the component of force along a displacement, token credit should reflect the local reward change along the policy update direction. We introduce Direction-Aware Token Credit Assignment (DATC), which maps frozen reward-model gradients into policy logit space and allocates weights by evaluating the alignment between the reward direction and GRPO's signed sampled-action update direction, while preserving the sign and mean of the response-level advantage. DATC outperforms GRPO in all four settings across two model families.
Oct 18, 2026

Outcome-Anchored Credit Reallocation for Agentic Knowledge Graph Question Answering

Agentic knowledge graph question answering (KGQA) requires large language models (LLMs) to solve complex questions through multi-turn interactions with a knowledge graph (KG). Outcome-supervised reinforcement learning derives verifiable outcome rewards from final answers but uniformly assigns each rollout's outcome-level advantage to all policy-generated tokens, creating a trajectory-to-operation mismatch since KG operations within a rollout can have varying effects on the search process and final answer. To address this, we propose Outcome-Anchored Credit Reallocation (OACR), which employs a training-only KG-state outcome model to estimate outcome-reward changes between adjacent KG search states and converts them into operation credit, from which token-wise weights are derived. These multiplicative weights reallocate token-level update magnitudes while preserving the sign of the outcome-level advantage at every generated token. Across three benchmarks, OACR achieves stronger overall performance than existing KGQA methods while using fewer inference-time LLM calls and tokens.
Oct 15, 2026

P2TPO: Structure-Aligned Page-to-Token Credit Assignment for Full-Page Document Parsing

End-to-end full-page document parsing must preserve the content and structure of heterogeneous elements, yet otherwise accurate outputs can contain local errors and omissions. A single page-level advantage does not explicitly identify which elements differ in quality across sampled outputs or where their errors occur. We introduce Page-to-Token Policy Optimization (P2TPO), a structure-aligned credit-assignment framework that operates on full-page policy trajectories. Reference-element alignment first associates block-level quality contrasts with generated spans. Unified Anchor Pricing (UAP) incorporates matched and missing element outcomes and provides feedback relative to a fixed quality ceiling when sampled outputs receive identical but imperfect scores for a reference element. Within matched blocks, Span Credit Redistribution (SCR) redistributes existing block credit according to aligned correct and erroneous spans, using boundary proxies for within-block omissions while preserving the total block-derived advantage across its tokens. Under a fixed-weight additive page utility, we show that element-indexed contrasts retain local quality differences that page-level aggregation can cancel out. In the main experimental comparison, P2TPO outperforms all evaluated policy optimization baselines in seed-averaged scores on each of the six document parsing benchmarks. Additional experiments demonstrate gains across model scales and initializations.
Oct 15, 2026

Enhancing Multimodal RLVR with Pairwise Token Contrast and Adaptive Trajectory Confidence

Reinforcement learning with verifiable rewards (RLVR) improves multimodal reasoning, but group-relative policy optimization broadcasts a single sequence-level advantage to every token of a response, although errors usually localize to a few decisive steps. Recent self-distilled variants refine credit with a teacher--student likelihood gap. However, this signal cannot tell whether it relies on the image or on a language shortcut, and is applied uniformly even when the teacher is uninformative. To address this, we propose Pairwise token contrast and Adaptive trajectory ConfidencE (PACE). Pairwise token contrast scores each response under four teacher views, crossing a correct or incorrect answer hint with the image present or removed, and compares the two hints for every token, which cancels shifts shared by both hints and reveals whether the token's support relies on the image. Adaptive trajectory confidence measures how well the teacher predicts each response given the correct hint and the image, normalizes this score within the batch, and uses it to set how far the token contrast redistributes credit within that response. PACE achieves the best average accuracy on three backbones and ranks first on all evaluated multimodal reasoning benchmarks with Qwen3-VL-8B-Instruct. Experiments show that both components matter: replacing pairwise token contrast with a single-view signal underperforms our method, while adaptive trajectory confidence adds further improvement.
Oct 15, 2026

Each Reward to Its Relevant Tokens: Reward-Specific Token-Level Credit Assignment for Multi-Reward RL

Language models are increasingly trained against several verifiable rewards at once, such as the correctness and the format of a tool call, or the individual constraints of a complex instruction. Reward-decoupled normalization (GDPO) normalizes each reward separately so that different reward combinations keep distinct advantages, but it still spreads every reward's advantage evenly over the whole response, although each verifier checks only a few positions. We propose MRTPO, which assigns the advantage of each reward to the tokens relevant to it, using the information its verifier checks, without auxiliary models or token-level annotations. For each reward, the policy replays its own response with and without that reward's verifier information, and the per-token Jensen-Shannon distance between the two next-token distributions measures each token's relevance and reassigns the advantage. On text-to-SQL, multi-constraint instruction following and tool calling, MRTPO converges faster and achieves better performance than the compared methods.
Oct 13, 2026

Unlocking Token Rewards via Training-Free Reward Attribution

In this paper, we propose an extremely efficient, training-free method to extract token-level reward signals directly from an existing deep reward model. Our core idea is to attribute the overall process reward to individual tokens by estimating each token's influence. This influence is defined as the change in the final macroscopic reward (e.g., the process reward) when a token is replaced with a semantically null token. Naively calculating this influence is computationally infeasible, requiring N forward passes through the PRM for an N-token sequence. We overcome this bottleneck by proposing a highly efficient gradient-based estimator. Specifically, we use a first-order Taylor approximation, which simplifies the influence calculation to the inner product of the difference between the token embedding and the null token embedding, and the gradient of the reward with respect to the token embedding. This requires only a single forward and backward pass. The resulting token-level rewards enable standard RL algorithms to perform precise credit assignment without requiring additional reward model training. Experiments on challenging reasoning benchmarks demonstrate that our method substantially improves policy optimization efficiency and enhances the generalization of LLM reasoning capabilities. Our P2T outperforms the outcome reward by +4.9% on MathVista for Qwen2.5-VL-7B-Instruct, and +11.5% on AIME24 for Qwen2.5-Math-7B. The code will be released via https://github.com/stonewst/P2T.
Oct 9, 2026

CAST: Credit Alignment for Cross-Tokenizer On-Policy Distillation

Cross-tokenizer on-policy distillation (OPD) requires aligning teacher and student sequences before their scores can be compared. Decoded-text alignment provides such correspondence at the chunk level, but leaves a distinct problem unresolved: credit alignment. A chunk-level signal specifies how much a span should change, but not how that signal should be distributed across student tokens, while tokens without teacher correspondence remain unconstrained. We propose CAST (Chunk-aware credit Assignment with Selective Token Anchoring), a shift-or-stay framework that addresses both issues. For aligned chunks, CAST redistributes each chunk's advantage across student tokens using predictive uncertainty and local teacher-signal smoothness, emphasizing learning opportunities supported by locally consistent teacher feedback while preserving the chunk's total advantage. The resulting weights exactly solve a KL-regularized credit-allocation problem. For unaligned positions, CAST selectively anchors the student to its initial policy through a KL penalty, limiting drift where teacher supervision is unavailable. Because credit assignment is separated from chunk construction, CAST also applies to punctuation-defined or fixed-length chunks under shared tokenization. Across six mathematics, science, and code benchmarks, CAST improves the average over the strongest evaluated baseline by 2.452.45 and 1.321.32 points for cross-tokenizer transfer from Qwen3-8B to Mistral-7B and Llama-3.1-8B, respectively, and by 2.102.10 points for shared-tokenizer transfer to Qwen3-1.7B.
Oct 6, 2026

VETTA: Coordinating Turn- and Token-Level Credit Assignment for Multi-Turn LLM Agents

Multi-turn LLM agents often receive sparse task feedback across several interactions, while generating each response token by token. This creates two related credit-assignment questions: which responses helped achieve the outcome, and which generation decisions mattered within each response? Existing methods typically focus on only one level: turn-level methods evaluate complete responses but do not distinguish the decisions within them; token-level methods can propagate feedback across turns but do not explicitly model credit for each response. These complementary limitations motivate learning credit at both levels and coordinating it in a single policy update. We introduce VETTA, a credit assignment method that jointly learns turn- and token-level values through separate heads on a shared lightweight critic. VETTA computes advantages along both temporal sequences and combines each turn advantage with a within-response-centered token residual for PPO updates. Furthermore, to reduce value-learning cost, the critic retains only early Transformer blocks from the pretrained checkpoint used to initialize the actor. Experiments on four agent benchmarks demonstrate VETTA's strong performance. With Qwen2.5-7B-Instruct, VETTA achieves success rates of 95.5% on ALFWorld, 76.0% on WebShop, and 52.2% on ScienceWorld. With Qwen2.5-VL-3B-Instruct, it reaches 84.9% and 80.7% success on Sokoban under sparse and dense rewards, respectively. Critic-depth comparisons further show strong task performance with substantially lower critic-side computation. These results demonstrate that coordinating turn- and token-level credit with a lightweight shared critic improves agent performance while keeping value estimation efficient. Code is available at https://github.com/Jiaju-Chen/VETTA-official.
Oct 2, 2026

Beyond Entropy: Self-Diagnostic Multi-Role Token Optimization for Video Reasoning

Reinforcement learning with verifiable rewards has substantially advanced multimodal reasoning, yet it remains fundamentally limited by ambiguous token-level credit assignment. While high-entropy token heuristics encourage possibility exploration, naively extending them to video reasoning tends to induce lengthy reasoning, as the model becomes overly reliant on high-entropy visual activations. Alternative approaches that rely on counterfactual-based visual token localization for credit assignment also tend to over-prioritize visual exploration at the expense of decisive reasoning cues for answer derivation, thereby exacerbating the interference from spurious visual nuances. Moreover, these methods employ static counterfactual strategies that fail to co-evolve with the policy during training. In this paper, we introduce DyCPO, a co-evolutionary framework that jointly optimizes reliable token selection and adaptive counterfactual intervention. It constructs a multi-role dependence metric to balance visual exploration and answer-relevance mining in token-wise contrastive learning, while suppressing exploration-only filler tokens and spurious visual noise. Rather than relying on static counterfactual priors, DyCPO dynamically derives counterfactual signals from the model's own successful and failed rollouts, enabling self-diagnostic analysis and co-evolution of the optimization objective with the policy. Extensive experiments on complex video reasoning and general video understanding benchmarks demonstrate consistent performance improvements, establishing DyCPO as a robust token-level credit assignment paradigm for multimodal reinforcement learning.
Oct 1, 2026

Token-Level Video Reinforcement Learning

Reinforcement learning (RL) for video generation usually assigns one scalar reward to an entire sampled video. Yet a video is not uniformly flawed: some visual tokens may already satisfy the prompt, whereas others require correction. A scalar reward cannot localize errors, causing optimization to perturb satisfactory tokens while under-targeting the tokens that actually need to change. We introduce Token-Level Video Reinforcement Learning, TVRL, a framework that derives token-level credit from the reward being optimized. Our key insight is that the answer likelihood of a frozen vision-language model provides both signals: its outputs contribute to the video-level reward, while magnitudes of its video-input gradients reveal which generated video tokens most affect that score. We instantiate TVRL in Group Relative Policy Optimization by averaging prompt-derived question rewards into one group-relative advantage and using detached, question-conditioned token-credit maps to reweight dense denoising-transition log-probabilities inside the clipped policy ratio. On VBench-2.0, TVRL achieves an Overall score of 57.69, outperforming the base model by 3.60 points. TVRL also improves matched GRPO baselines across three SDE samplers (SAGE, Flow, and Dance) by 2.68--3.15 points and across four reward models (VideoAlign, VideoScore2, UnifiedReward2, and Qwen3.5-9B) by 1.33--3.15 points.
Sep 30, 2026

Reinforcing Multimodal Reasoning via Token-Level Perception-Grounded Advantage Estimation

Reinforcement Learning with Verifiable Rewards (RLVR) has improved the reasoning capabilities of Multimodal Large Language Models (MLLMs), yet existing frameworks rely on coarse, sequence-level reward signals that lack the fine-grained supervision over the visually-grounded steps within a multimodal reasoning chain. We investigate this gap through the lens of two token-level metrics: visual dependency (i.e. how much a token's prediction relies on the input image features) and predictive entropy. Our empirical analysis reveals two key findings: (1) correct reasoning chains exhibit a markedly sharper entropy reduction as visual grounding intensifies, compared to incorrect ones; (2) pivotal tokens, those whose misprediction triggers reasoning collapse, are statistical outliers in the joint distribution of visual dependency and predictive entropy derived from correct chains. Motivated by these findings, we propose token-level perception-grounded advantage estimation (TPAE), which estimates token-level advantages by measuring each token's statistical consistency with the vision-entropy patterns of correct rollouts. TPAE leverages this granular score to modulate the sequence-level advantage, producing a fine-grained supervision signal that can be integrated into various RLVR frameworks. Extensive experiments on seven benchmarks show that TPAE consistently outperforms leading strong baselines, yielding more stable and efficient optimization for multimodal reasoning. The code is publicly available at https://github.com/Zhihan72/TPAE.
Sep 28, 2026

Can We Trust the Teacher? Decoupled Credit Direction-Magnitude for Self-Distillation

RLVR provides reliable trajectory-level credit, while OPSD offers dense supervision for token-level credit. This exposes a fundamental coupling when updating step-level credit direction and magnitude with teacher supervision, preventing steps from receiving reliable credit directions and contribution magnitudes, while making both vulnerable to teacher judgment errors and preference variance, as supported by our theoretical analysis. To separate credit direction from its contribution magnitude, we introduce Decoupled Credit Self-Distillation (DCSD), which theoretically decouples credit direction and magnitude into two reliable signals and uses them to calibrate privileged teacher supervision. Specifically, we design belief-margin probing to determine credit direction and marginal information gain to quantify credit magnitude, enabling step-to-token credit assignment for policy optimization. Across 11 benchmarks, DCSD achieves the best overall scores against GRPO, OPSD, RLSD, and RLCSD. Compared with base models, DCSD improves the overall score by 8.45 points on mathematical reasoning and 7.01 points on multimodal reasoning, while correcting the credit direction for 6% of tokens and yielding a 1.5×\times reduction in token credit magnitude.
Sep 22, 2026

PACT: From Credit Assignment to Critic Alignment

Reinforcement learning has become a central component of large language model (LLM) post-training, yet token-level credit lacks a generally accepted mathematical definition, leaving its relationship to commonly used training signals unclear. We formulate three regularity conditions, namely Completeness, Prefix Consistency, and Neutrality, and prove that they uniquely determine token-level credit. This characterization provides a unified basis for explaining phenomena across existing algorithms and guides the development of an improved actor-critic training procedure. Through this lens, an ideal teacher in On-Policy Distillation (OPD) acts as an implicit critic, yielding an expected policy gradient proportional to that induced by token-level credit. Response-level REINFORCE Leave-One-Out (RLOO) signals match the expected policy-gradient contribution of token-level credit despite their coarser granularity. We further establish approximate credit sparsity under bounded outcome rewards and show how intermediate critic errors in Generalized Advantage Estimation (GAE) can become comparable to the underlying credit. These motivate Policy Aligned Critic Training (PACT), which adopts an Actor-then-Critic update order to apply importance sampling correction to critic training and better align the critic with the updated policy. In agentic mathematical reasoning, PACT achieves 72.87% average accuracy across four benchmarks, outperforming GRPO and PPO by 8.80 and 13.16 percentage points, respectively. On SWE-bench Verified, PACT achieves a pass rate of 67.4%, outperforming PPO, GRPO, and SAO by 2.4, 2.0, and 3.8 percentage points, respectively.
Sep 15, 2026

TIAO: Token Importance-Aware Policy Optimization for Text Summarization

Text summarization requires models to condense content while preserving key qualities such as consistency and coherence. Large language models (LLMs) have shown strong performance on this task and can be further improved through reinforcement learning (RL). However, most existing methods apply reward signals directly to undifferentiated token sequences, overlooking the varying importance of individual tokens to word and sentence level quality in summarization. In this paper, we propose Token Importance-Aware Policy Optimization (TIAO), a novel reinforcement learning strategy that explicitly leverages token-importance awareness. Specifically, TIAO identifies core tokens based on token dependency and reweights a trajectory's advantage according to its overall dependencies. Experiments on the real world dataset show that our TIAO achieves highly competitive results, and that a 7B foundation model enhanced by TIAO performs comparably to GPT-4 and GPT-5-nano. Code is available at https://github.com/TechCloud-x/TIAO
Aug 21, 2026

Let Credit Follow Computation: Architecture-Aware Credit Transport for Large Language Model Reinforcement Learning

Credit assignment in large-language-model reinforcement learning (LLM RL) can be separated into three objects: evidence about success, a transport operator that converts this evidence into token-level advantages, and an update geometry that turns advantages into policy changes. Recent work has greatly improved evidence, sampling, and update geometry, but the transport operator is usually architecture-agnostic. Fixed-discount GAE applies a stationary geometric kernel along token time; group-relative methods broadcast an outcome statistic across an entire response. Neither operator represents the trajectory-specific computation used by the Transformer policy itself. We introduce computation-conditioned credit transport (CCT), a general framework in which a detached statistic of the behavior policy's internal computation parameterizes the causal kernel that transports downstream value through a rollout. Our concrete algorithm, CompPO, maps native attention concentration to a bounded per-token retention gate, uses the gate in both the one-step bootstrap and a path-dependent generalized-advantage trace (Comp-GAE), and co-designs a transport-aligned critic (TAC) that reuses the actor's hidden states and routing information without a second same-scale Transformer. The task reward and clipped PPO policy objective remain unchanged; a constant gate recovers fixed-coefficient GAE. Across five Qwen3-4B seeds, CompPO reaches 61.4% final held-out accuracy (95% CI [60.8,62.0]) versus 53.8% [52.9,54.7] for tuned GRPO. Neither Comp-GAE with a standard critic (55.2%) nor TAC with a fixed gate (56.4%) matches the full model (interaction +2.4 [1.9,2.9]). Shuffle and position controls confirm trajectory-specific alignment; CompPO is stable in 10/12 PPO-grid runs versus 3/12. Frozen evaluation improves over GRPO by 4.3 and 3.9 greedy pass@1 macro points on Qwen3-4B and Llama-3.1-8B-Instruct.
Aug 19, 2026

Beyond Teacher Likelihood: Group-Calibrated On-Policy Distillation for Long-Context Reasoning

On-policy distillation (OPD) trains a student on its own responses using dense token-level guidance from a stronger teacher. In long-context tasks, however, token-level teacher support can favor locally plausible responses that omit evidence distributed across the input or violate global task constraints. Task-specific verifiers, in contrast, evaluate task completion at the response level and may return graded rewards that reflect partial success. We diagnose this mismatch on fixed responses from two representative long-context evidence-aggregation tasks. Across longer input ranges, trajectory-level OPD scores become progressively less aligned with verifier rewards, indicating teacher-verifier disagreement. Motivated by this observation, we introduce Group-Calibrated On-Policy Distillation (GC-OPD). GC-OPD separately normalizes verifier rewards and trajectory-level OPD scores within each rollout group and uses their difference as a signed teacher-verifier disagreement residual. Relative-advantage-based credit assignment (RACA) distributes this trajectory-level residual across tokens according to their relative OPD advantages while preserving the original OPD signal. Across five long-context benchmarks, post-training with GC-OPD raises the five-benchmark averages of the official Qwen3-4B and Qwen3-8B checkpoints from 29.08 to 40.47 and from 35.12 to 44.65, respectively. Vanilla OPD reaches 39.31 and 43.56 under the same setup. Controlled ablations show that the signed residual is more effective than either an additional OPD-derived term or direct group-normalized verifier reward addition, while RACA further improves over uniform token allocation. Together, these results demonstrate that group-relative residual calibration can incorporate verifier outcomes without discarding dense token-level guidance. Code is available at https://github.com/SolereZhang/GC-OPD.
Aug 17, 2026

Le Critique: Privileged Value Functions for LLM Reinforcement Learning

Reinforcement learning algorithms for Large Language Models (LLMs) are largely distinguished by their variance reduction strategy. Group-relative methods like GRPO reduce gradient variance by sampling multiple rollouts per prompt, but provide only sequence-level credit. Training is also blocked by straggler rollouts, reducing throughput and increasing off-policyness. Learned value functions theoretically address both problems, providing token-level advantages without requiring large groups. However, additional infrastructure engineering challenges combined with the practical success of critic-free methods have made it difficult to justify their inclusion in RL pipelines. We propose two complementary strategies to improve the performance of value function RL: 1) Privileged Value Functions (PVF) which provide an elegant mechanism to inject additional task-relevant token-level signal without biasing the policy objective; 2) TETHER, a baseline that adaptively interpolates between group-relative and value baselines depending on the value function accuracy. Across several reasoning tasks, both strategies consistently improve over the standard value function baseline, and are competitive with or outperform mean-baseline GRPO.
Aug 13, 2026

Teach the Magnitude, Not the Direction: Verifier-Bounded Credit Assignment for Multi-Turn Multi-step LLM Agents

Reinforcement learning with verifiable rewards (RLVR) offers a verifier-bounded performance ceiling for training multi-turn tool-use agents, yet its trajectory-level credit assignment conflates heterogeneous per-turn outcomes into a single reward signal. On-policy distillation provides dense per-token supervision but is either teacher-bounded or prone to gradient concentration collapse. We introduce CrEST\textbf{CrEST}, a hierarchical credit assignment framework that retains RL's verifier-bounded ceiling while incorporating dense token-level signals from a privileged self-teacher. CrEST\textbf{CrEST} resolves credit at two levels: turn-segmented verified advantages address inter-turn dilution, while entropy-gated self-teacher modulation refines intra-turn token contributions. Experiments on BFCL V3 and WildToolBench show that CrEST\textbf{CrEST} consistently outperforms both RL and distillation baselines across two model scales, with the largest gains on long-trajectory and strict session-level metrics. Our work demonstrates that the teacher's role in policy optimization can be reduced from determining update directions to modulating update magnitudes, unlocking dense credit assignment without sacrificing the verifier-bounded ceiling.
Aug 10, 2026

Se-DPO: Self-Evolving Token Credit for Direct Preference Optimization

Direct Preference Optimization (DPO) aggregates token-level log-probability ratios via uniform summation, implicitly treating all tokens as contributing equally to the preference signal. However, the contribution of individual tokens to the preference signal varies. We introduce token credit, which modulates each token's KL regularization based on its contribution to the preference outcome. We derive that effective token credit is proportional to the magnitude of each token's implicit reward, and observe that this quantity evolves substantially during training. This implies that static token credit becomes increasingly misaligned as training progresses. In this work, we propose Se-DPO (Self-Evolving Token Credit for DPO), a live mechanism that derives token credit from the model's own evolving internal signals during DPO training. Since the reward signal varies in reliability across positions, Se-DPO calibrates token credit based on both the strength and the confidence of each token's contribution. Se-DPO requires no external models, adding only a lightweight calibration network with minimal computational overhead. Experiments show that Se-DPO improves over DPO by up to 9.8 points on AlpacaEval~2 and 12.2 points on Arena-Hard.
Aug 10, 2026

Privileged Likelihood Is Not Automatically Value: Three Checks for Token Credit in On-Policy Self-Distillation

On-policy self-distillation aims to improve upon reinforcement learning from verifiable rewards (RLVR) by providing token-level scores derived from privileged information, such as reference solutions or critic feedback. These scores are treated as estimates of token-level action values, yet they answer a fundamentally different question: how the model's prediction changes when its input context is enriched, rather than how the expected outcome changes when a token is changed. We examine this gap along three dimensions: (i) whether the token-level score tracks task success; (ii) whether feedback generated from the same rollout causes the score to reflect agreement with its own description, and whether using feedback from other rollouts in the group mitigates this self-referential effect; and (iii) what behavior the resulting training objective actually reinforces. In experiments on AIME 2025, the implemented score distinguishes correct from incorrect rollouts at approximately chance level (AUC=0.505); using feedback from a different rollout does not consistently improve this discrimination; and all training configurations achieve only 24.2-33.9% Avg@4, compared with 64.2% for outcome-only GRPO. Moreover, the highest-entropy token decile accounts for 57-71% of the total absolute token-advantage mass, despite the score being least informative about reasoning quality in this regime. By contrast, similar experiments on SciKnowEval Biology improves held-out Avg@8 by 28.0%, while its corresponding trajectory scores achieve AUCs of 0.81-0.92. Together, these results suggest that dense credit assignment through distillation can be effective when its likelihood-based scores are empirically validated as meaningful proxies for outcome-relevant credit. When this alignment does not hold, however, the resulting supervision can fail to generalize and may substantially underperform outcome-based RL.
Aug 7, 2026

How Much, Then Where: Credit-Conserving Action-to-Token Allocation for Multi-Turn Agent Reinforcement Learning

Credit assignment in multi-turn agent reinforcement learning operates at two levels: assigning trajectory-level credit to actions and distributing each action's credit across its tokens. In this paper, we introduce FACTOR, which separates these decisions. FACTOR uses checkpoint-calibrated TD residuals to assign per-action credits that telescope to the trajectory advantage, and feedback-conditioned teacher-student likelihood gaps to allocate each credit across the realized action tokens. Per-action normalization preserves the action-average coefficient and prevents token-level sign flips. We pair this construction with an action-mean reduction, removing the implicit dependence of an action's scalar surrogate weight on its token length. At the behavior policy and before clipping, each action's inner action-mean surrogate equals its TD credit. FACTOR consistently improves over competitive baselines across ALFWorld, WebShop, and ScienceWorld, with every environment-seed comparison favoring FACTOR and the largest gains emerging on the longest-horizon environment. The same hyperparameters transfer without retuning to a larger backbone and to a different model family. Ablations identify TD action credit as the dominant driver of the improvement, with hindsight token allocation contributing complementary gains.
Aug 4, 2026

Agentic Reinforcement Learning with Self-Distilled Reward Shaping

Agentic reinforcement learning enables LLM agents to learn through interaction, but sparse trajectory-level rewards reveal success without identifying which intermediate decisions deserve credit. Training-only privileged skills can provide denser supervision by allowing the same frozen policy snapshot to rescore fixed tokens from skill-free trajectories while conditioned on task-matched procedural skills. Existing methods, however, do not jointly calibrate teacher scores across interaction steps, relate teacher confidence to realized returns, and integrate the resulting signal into native reward-to-advantage construction. We introduce Agentic Reinforcement Learning with Self-Distilled Reward Shaping (ADRS), a framework for constructing return-associated token-level credit for multi-turn language agents. ADRS centers and normalizes privileged token scores within each step, modulates them with a return-associated Teacher Value Advantage (TVA) gate based on within-group confidence--return association, and incorporates the gated token signal into native RL credit construction. Together, these components determine what the teacher prefers, when that preference is return-relevant, and how it enters the native reinforcement-learning credit path, while keeping rollouts and inference skill-free. Finally, experiments across three interactive benchmarks show that ADRS consistently improves performance on long-horizon tasks, with gains persisting across RL backbones, reduced-data settings, unseen tasks, and extended training. For anonymous review, our code is available at the following the link: https://github.com/gitrxh/ADRS-arxiv
Jul 30, 2026

Not All Tokens Deserve Equal Credit: Counterfactual Sensitivity Credit Reallocation for Long-CoT Reasoning

Reinforcement learning with verifiable rewards (RLVR) is central to improving long-CoT reasoning in large language models. Critic-free methods such as GRPO convert response-level rewards into advantages and uniformly broadcast them across tokens, overlooking their unequal contributions to the final outcome. On-policy self-distillation (OPSD) instead provides dense distributional supervision by minimizing the forward KL divergence between an unprivileged policy and a privileged self-teacher, implicitly assuming that the resulting likelihood shifts encode reliable answer-aligned information. We test this premise by fixing each sampled trajectory and re-scoring it under two opposing outcome conditions, one asserting correctness and the other incorrectness. Most affected tokens shift in the same direction under both conditions, with few sign reversals and substantial overlap in the induced optimization signals. Large shifts also concentrate on highly substitutable surface-form tokens, whereas tokens carrying problem-specific reasoning content are less sensitive. These findings show that privileged shifts fail to provide reliable answer-aligned directions, while their magnitudes primarily reflect counterfactual sensitivity rather than token-level learning value. Based on these observations, we propose Counterfactual Sensitivity Credit Reallocation (CSCR), a simple extension of GRPO that reduces credit for highly sensitive tokens and renormalizes token-level advantages to preserve both the original credit budget and verifier-determined direction. On long-CoT mathematical reasoning benchmarks, CSCR consistently outperforms GRPO baseline with the same number of policy updates. Targeted ablations further corroborate our diagnosis: privilege-induced directions are unreliable, moderate downweighting is most effective, and stronger modulation destabilizes optimization.
Jul 28, 2026

CoRT: Counterfactual Replay for Token-Level Rubric-Guided Policy Optimization

Rubric-based reinforcement learning enriches language model training by evaluating model outputs against explicit criteria. Yet in GRPO-style pipelines, these structured judgments are reduced to a scalar response-level reward and converted into a response-level advantage, which is broadcast uniformly to all generated tokens. This leaves no explicit mechanism for allocating credit within a response, even when different criteria are grounded in different spans, formatting decisions, or semantic choices. We propose CoRT, a token-level credit weighting method for rubric-conditioned GRPO. Instead of training an auxiliary token scoring model, CoRT uses counterfactual replay to rescore the same sampled response under the original rubric-conditioned prompt and a matched criteria-free prompt. The resulting tokenwise log-likelihood contrasts serve as a proxy for dependence on the rubric context. CoRT maps these contrasts to bounded, response-normalized weights and uses them to redistribute the signed GRPO advantage across tokens, without introducing an auxiliary scorer or changing the response-level reward. Experiments across instruction-tuned models and reward granularities show that CoRT improves over matched response-level GRPO in the vast majority of comparisons, with an average gain of 4.4 percentage points. The method remains competitive with learned token-level credit baselines while avoiding a separate relevance-learning stage. These results suggest that policy-internal counterfactual likelihood contrasts provide an effective training signal for within-response credit allocation while retaining the simplicity and stability of GRPO.
Jul 8, 2026

Right Answer, Wrong Tokens: Tail-Aware Credit Calibration for RLVR

Although reinforcement learning (RL) improves the reasoning capabilities of large language models (LLMs), GRPO-style methods can suffer from a positive-credit contamination failure mode. Specifically, uniform credit assignment can reinforce locally erroneous tail tokens alongside useful tokens within successful trajectories. To address this problem, we propose Tail-Aware Credit calibratiOn (TACO), which calibrates the positive credit assigned to such tokens without token-level error labels. To this end, TACO lets the policy serve as its own reference, assessing each sampled token's rarity relative to the policy's local expectations to compute a tail-risk score. Using this score as a proxy for potential local errors, TACO softly reduces positive credit for high tail-risk tokens while retaining nonzero weights, limiting their reinforcement without eliminating opportunities to learn useful rare patterns. Experiments across four LLMs and eight reasoning benchmarks show that TACO consistently improves over GRPO-style baselines. Notably, TACO also improves training stability, supporting sustained performance gains in long-horizon RL. The source code is available at: https://github.com/xiuyilou/TACO.