SEVA: Self-Evolving Verification Agent with Process Reward for Fact Attribution

Aojie Yuan, Yi Nian, Haiyue Zhang, Zijian Su, Yue Zhao

• PDF
• arXiv
• DOI
• Explainer
• 中文解读

Abstract

Hallucination is the reliability bottleneck for LLM-based agents, and fact attribution verifiers are the last line of defense -- yet today's verifiers emit only opaque binary labels, leaving agents unable to self-correct and operators unable to audit. We present SEVA, a structured verification agent that emits evidence alignments, step-by-step reasoning chains, calibrated confidence, and a six-category error diagnosis with actionable fixes. Training such an agent with RL is non-trivial: standard binary reward on multi-component output triggers advantage collapse -- within-group reward variance vanishes and the GRPO gradient disappears. We resolve this with a process reward that decomposes verification quality into five independent components weighted 70/30 toward process signals, restoring the gradient and inducing an implicit curriculum -- the agent first masters verification behavior (alignment 0.917 -> 0.997, format 72% -> 100%), then outcomes (F1 64.9 -> 69.0). Structured output further enables a Verify -> Reflect -> Probe -> Refine self-evolution loop, which over four rounds on a 7B model surfaces an unexpected structural finding: each round produces a benchmark-specialist, not a generalist (+15 pp on HaluEval, -10 to -14 pp on TruthfulQA in the same model, persistent at 4x data). On ClearFacts, SEVA-3B matches GPT-4o-mini (69.0 vs. 69.8 F1) while producing substantially richer, auditable output -- confirming a principle that should generalize: for any RL task with multi-component generation, reward granularity must match output granularity.

Citation

@inproceedings{yuan2026seva,
  title     = {SEVA: Self-Evolving Verification Agent with Process Reward for Fact Attribution},
  author    = {Yuan, Aojie and Nian, Yi and Zhang, Haiyue and Su, Zijian and Zhao, Yue},
  booktitle = {AI4GOOD and FAGEN Workshops at ICML 2026},
  year      = {2026},
  eprint    = {2606.29713},
  archivePrefix = {arXiv},
  doi       = {10.48550/arXiv.2606.29713},
  url       = {https://arxiv.org/abs/2606.29713}
}