HindSearch: Trajectory-Level Hindsight Critique for Search-Augmented Reinforcement Learning
Mirrored from arXiv — Machine Learning for archival readability. Support the source by reading on the original site.
Computer Science > Machine Learning
Title:HindSearch: Trajectory-Level Hindsight Critique for Search-Augmented Reinforcement Learning
Abstract:Search-augmented LM agents are typically trained with a binary exact-match reward, which throws away most of what a failed trajectory tells us about why it failed. We introduce HindSearch, a hindsight self-distillation procedure for GRPO: after each rollout, a frozen judge writes a short critique of every failed trajectory using the gold answer, and the critique supplies an auxiliary on-policy distillation signal on the student's search actions. On the standard seven-benchmark suite with Qwen2.5-3B-Instruct, HindSearch reaches 39.4% average EM, outperforming prior search-RL baselines. Removing the judge's access to the gold answer erases most of the gain, isolating hindsight as the source of the improvement.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR) |
| Cite as: | arXiv:2608.01597 [cs.LG] |
| (or arXiv:2608.01597v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2608.01597
arXiv-issued DOI via DataCite (pending registration)
|
Access Paper:
- View PDF
- HTML (experimental)
- TeX Source
Current browse context:
References & Citations
Bookmark
Bibliographic and Citation Tools
Code, Data and Media Associated with this Article
Demos
Recommenders and Search Tools
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
More from arXiv — Machine Learning
-
MS-MLB: An Open Machine Learning Benchmark for Blood-Based MS Classification
Aug 7
-
When Do Corrective Features Help? An Agent for Corrective Feature Discovery on Black-Box Forecasters
Aug 7
-
PPDL: LLM-Based Flows as Probabilistic Programs
Aug 7
-
Decoupling Perception from Description: Computation-Grounded Representation Alignment between Multivariate Time Series and Language
Aug 7
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.