EviSD: Evidence-Conditioned Self-Distillation for Search-Augmented Agents

• URL: http://arxiv.org/abs/2608.01359v1
• Date: Sun, 02 Aug 2026 16:22:27 GMT
• Title: EviSD: Evidence-Conditioned Self-Distillation for Search-Augmented Agents
• Abstract summary: We propose EviSD, an evidence-conditioned self-distillation framework.<n>It uses instance-level supporting evidence as privileged information for search actions and golden answers.<n>EviSD achieves the highest macro-average Exact Match in all evaluated settings.
• Score: 28.95182429824442
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Outcome-based reinforcement learning enables search-augmented language agents to learn from verifiable final answers, but its trajectory-level credit cannot distinguish the contributions of individual actions in a multi-turn search process. We propose EviSD, an evidence-conditioned self-distillation framework that uses instance-level supporting evidence as privileged information for search actions and golden answers as complementary privilege for answer actions. During training, the student samples actions from the original context, while the same model re-scores them as a privileged teacher under an action-aligned context. EviSD converts the detached teacher--student gap into a bounded correction to the outcome-derived GRPO advantage and applies it only to generated action spans. This design localizes privileged guidance while preserving the update direction determined by the outcome reward, without an auxiliary distillation objective or any change at inference time. Across seven question-answering benchmarks and three backbones spanning model scales and generations, EviSD achieves the highest macro-average Exact Match in all evaluated settings, outperforming the strongest compared methods by 1.3--2.3 points while modulating only 6.7%--15.1% of response tokens. Code is available at https://github.com/JiananXie/EviSD.

Related papers

• Beyond Outcome Rewards: Step-Level Self-Distilled Policy Optimization for Deep Search Agents [15.62688304085298]
  Deep search agents operate over trajectories spanning dozens of steps.<n>On-policy self-distillation (OPSD) addresses this by using the model's own logits as dense token-level teachers.<n>We propose Step-Level Self-Distilled Policy Optimization (SSPO), which converts teacher-student disagreement into step-level advantage weights.
  arXiv  Detail & Related papers  (2026-08-13T03:13:18Z)
• Distill Where You Fail: Recovering Learning Signals of Negative RL-Groups from Adaptive Teacher Guidance [50.77757355236912]
  We propose RSTG (Recovering Learning Signals via Adaptive Teacher Guidance), which applies distillation selectively and precisely where it matters most.<n> Experiments demonstrate that RSTG substantially outperforms naive GRPO+OPD by +4.02% on math and+3.05% on code.
  arXiv  Detail & Related papers  (2026-08-01T17:29:14Z)
• EviBack: Search-Agent Reinforcement Learning via Evidence-Constrained Teacher Backoff [12.413369909547699]
  EviBack is an evidence- constrained Teacher backoff that supplies auxiliary super- vision to such groups.<n>A fully automated, end-to-end GPT- 5.5-assisted APE pipeline starts from a manually authored single-prompt dual-task Teacher.
  arXiv  Detail & Related papers  (2026-07-27T02:59:27Z)
• H$^2$SD: Hybrid Hindsight Self-Distillation [61.71131241473028]
  Reinforcement learning with verifiable rewards (RLVR) provides reliable outcome supervision for language model reasoning.<n>Existing self-distillation methods add a privileged teacher but typically assign it a fixed role.<n>We introduce Hybrid Hindsight Self-Distillation ($mathrmH2mathrmSD$), which jointly adapts teacher context and update strategy to trajectory correctness.
  arXiv  Detail & Related papers  (2026-07-21T10:47:27Z)
• LOTAPO: Leave-One-Turn Attribution for Self-Generated Process Rewards in Multi-Turn Search Reasoning [12.013691798765594]
  LOTAPO is a self-generated process-supervision method based on backward leave-one-turn attribution.<n>For each search turn, LOTAPO replaces the turn and its retrieval observation with a fixed [DELETE] placeholder.<n>This Answer-Likelihood Gain estimates the turn's contribution while preserving all downstream interactions.
  arXiv  Detail & Related papers  (2026-07-15T06:55:28Z)
• Rethinking Reward Supervision: Rubric-Conditioned Self-Distillation [60.55792673956761]
  We propose textbfRubric-Conditioned Self-Distillation, a framework that incorporates rubrics as structured, fine-grained feedback for on-policy self-distillation.<n>We evaluate on a diverse suite of science reasoning benchmarks and results show that rubric-conditioned self-distillation effectively converts rubric-level criteria into token-level guidance.
  arXiv  Detail & Related papers  (2026-06-17T17:54:04Z)
• Localizing Credit at the Divergence: Path-Conditioned Self-Distillation for LLM Reasoning [12.363923974156636]
  Reinforcement learning from verifiable rewards assigns a single scalar to each rollout.<n>On-policy self-distillation addresses this by letting the same model act as a teacher conditioned on privileged information.<n>We propose Hindsight Self-Distillation (HSD), which conditions the teacher on a successful peer rollout.
  arXiv  Detail & Related papers  (2026-06-14T03:37:27Z)
• HERO: Hindsight-Enhanced Reflection from Environment Observations for Agentic Self-Distillation [50.53459634301361]
  HERO is a hindsight-enhanced self-distillation framework that uses next environment observations as locally aligned feedback.<n> HERO improves task success and reduces unnecessary turns over environment-feedback-only self-distillation and GRPO.
  arXiv  Detail & Related papers  (2026-06-10T01:35:34Z)
• VeriGate: Verifier-Gated Step-Level Supervision for GRPO [51.26100506256885]
  Group Relative Policy Optimization is an effective recipe for training reasoning models with verifier-based outcome rewards.<n>We present VeriGate, a verifier-gated extension of GRPO that addresses these limitations with three design choices.<n>We show that VeriGate improves average accuracy by about 20% and 12% for 1.5B and 7B models respectively.
  arXiv  Detail & Related papers  (2026-05-28T18:20:32Z)
• OGLS-SD: On-Policy Self-Distillation with Outcome-Guided Logit Steering for LLM Reasoning [19.98950359294245]
  We study on-policy self-distillation (OPSD), where a language model improves its reasoning ability by distilling privileged teacher distributions along its own on-policy trajectories.<n>Despite the performance gains of OPSD, we identify a common but often overlooked mismatch between teacher and student responses.
  arXiv  Detail & Related papers  (2026-05-12T17:00:53Z)
• Look As You Think: Unifying Reasoning and Visual Evidence Attribution for Verifiable Document RAG via Reinforcement Learning [55.232400251303794]
  Look As You Think (LAT) is a reinforcement learning framework that trains models to produce verifiable reasoning paths with consistent attribution.<n>LAT consistently improves the vanilla model in both single- and multi-image settings, yielding average gains of 8.23% in soft exact match (EM) and 47.0% in IoU@0.5.
  arXiv  Detail & Related papers  (2025-11-15T02:50:23Z)

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.