# DRACO: Fine-Grained Credit Assignment with Dynamic Rubrics for Long-Horizon Agent Training

arXiv: https://arxiv.org/abs/2609.04094 · HTML: https://mlsanity.com/paper/2609.04094/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 50 references that are in the map.
- First posted on arXiv 3 Sep 2026.

## Builds on (most cited first, 28 of 28)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [Group-in-Group Policy Optimization for LLM Agent Training](https://mlsanity.com/paper/2505.10978/) (2025)
- [AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agents](https://mlsanity.com/paper/2407.18901/) (2024)
- [Spurious Rewards: Rethinking Training Signals in RLVR](https://mlsanity.com/paper/2506.10947/) (2025)
- [RUDDER: Return Decomposition for Delayed Rewards](https://mlsanity.com/paper/1806.07857/) (2018)
- [DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research](https://mlsanity.com/paper/2511.19399/) (2025)
- [Reinforcement Learning for Long-Horizon Interactive LLM Agents](https://mlsanity.com/paper/2502.01600/) (2025)
- [OpenClaw-RL: Train Any Agent Simply by Talking](https://mlsanity.com/paper/2603.10165/) (2026)
- [AgentEvolver: Towards Efficient Self-Evolving Agent System](https://mlsanity.com/paper/2511.10395/) (2025)
- [Hindsight Credit Assignment](https://mlsanity.com/paper/1912.02503/) (2019)
- [SPA-RL: Reinforcing LLM Agents via Stepwise Progress Attribution](https://mlsanity.com/paper/2505.20732/) (2025)
- [Hindsight Credit Assignment for Long-Horizon LLM Agents](https://mlsanity.com/paper/2603.08754/) (2026)
- [Alternating Reinforcement Learning for Rubric-Based Reward Modeling in Non-Verifiable LLM Post-Training](https://mlsanity.com/paper/2602.01511/) (2026)
- [Chasing the Tail: Effective Rubric-based Reward Modeling for Large Language Model Post-Training](https://mlsanity.com/paper/2509.21500/) (2025)
- [Online Rubrics Elicitation from Pairwise Comparisons](https://mlsanity.com/paper/2510.07284/) (2025)
- [SALT: Step-level Advantage Assignment for Long-horizon Agents via Trajectory Graph](https://mlsanity.com/paper/2510.20022/) (2025)
- [Reliability without Validity: A Systematic, Large-Scale Evaluation of LLM-as-a-Judge Models Across Agreement, Consistency, and Bias](https://mlsanity.com/paper/2606.19544/) (2026)
- [StepTool: Enhancing Multi-Step Tool Usage in LLMs via Step-Grained Reinforcement Learning](https://mlsanity.com/paper/2410.07745/) (2024)
- [SCRIBE: Structured Mid-Level Supervision for Tool-Using Language Models](https://mlsanity.com/paper/2601.03555/) (2026)
- [TRACE: Turn-level Reward Assignment via Credit Estimation for Long-Horizon Agents](https://mlsanity.com/paper/2607.13988/) (2026)
- [RubricEM: Meta-RL with Rubric-guided Policy Decomposition beyond Verifiable Rewards](https://mlsanity.com/paper/2605.10899/) (2026)
- [EvoRubric: Self-Evolving Rubric-Driven RL for Open-Ended Generation](https://mlsanity.com/paper/2605.29847/) (2026)
- [Rubrics to Tokens: Bridging Response-level Rubrics and Token-level Rewards in Instruction Following Tasks](https://mlsanity.com/paper/2604.02795/) (2026)
- [The Verification Horizon: No Silver Bullet for Coding Agent Rewards](https://mlsanity.com/paper/2606.26300/) (2026)
- [Discretizing Reward Models](https://mlsanity.com/paper/2606.21795/) (2026)
- [Learning CLI Agents with Structured Action Credit under Selective Observation](https://mlsanity.com/paper/2605.08013/) (2026)

## Built on by (most cited first, 5 of 5)

- [CREDO: Variance-Guided Rubric Evolution for Replay-Corrected Credit Assignment](https://mlsanity.com/paper/2609.24174/) (2026)
- [Matched-Context Group Policy Optimization for LLM Agent Training](https://mlsanity.com/paper/2610.13755/) (2026)
- [Of Errors and Echoes: Responsibility-Aware Credit Assignment for Tool-Use Agents](https://mlsanity.com/paper/2610.16027/) (2026)
- [CoCA: Criterion-Aware Credit Assignment for Rubric-Based Reinforcement Learning](https://mlsanity.com/paper/2610.17733/) (2026)
- [WorkRover: Scaling Long-Horizon Cowork Agent Training Via Recursive Environment Evolution](https://mlsanity.com/paper/2610.18067/) (2026)

## Abstract (arXiv)

Reinforcement Learning from Verifiable Rewards works well when a task has a programmatic checker, but most long-horizon agent domains have none. We work in the outcome-blind setting, where ground-truth success signals are not available. Multi-criteria rubrics are a popular way to supply such a reward; they are scored once per trajectory, but a single scalar is a poor signal across tens of steps. We propose DRACO: Distributing Rubric-based Advantage for Credit Optimization. It generates rubrics dynamically during training to track the policy's evolving capability, scores those rubrics once per completed trajectory, and redistributes that judgment over the steps responsible for annotated rubrics to produce differentiated per-step advantages in GRPO. The redistribution is closed-form and does not introduce any trained attribution module. On AppWorld, DRACO gains 15.9 points over the base model and 5.3 points over GRPO trained with a sparse ground-truth reward, despite not using any verifiers itself. On out-of-domain Tau-Bench, it gains 5.3 points over the base model even without a frontier judge, beating both ground-truth-reward training and other rubric-based training settings. The code for DRACO is available at https://github.com/IBM/draco.