CLIFT: Conformal Self-Verification for Web Agent Training and Test-Time Scaling
Open-source web agents are now strong enough to execute realistic browser tasks, but training them with reinforcement learning still depends on weak supervision: binary task success is too sparse for credit assignment, while frontier-language-model judges are too expensive to call at every step and cannot be assumed available at deployment. We introduce CLIFT, a training and test-time scaling method built around conformal self-verification. During training, the agent answers natural-language verification questions about its own rollouts; a Compositional Conformal Certifier keeps only question signals whose URL-conditional evidence agrees with a training-time judge, assigns signed trust we...
| Authors | Yifan Zhang, Yutong Dai, Viraj Prabhu et al. |
|---|---|
| Primary category | cs.CL |
| Published | 5 October 2026 |
| Last updated at source | 5 October 2026 |
Popularity, last 2 days
0 now. Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 16 Oct 2026 Description: Open-source web agents are now strong enough to execute realistic browser tasks, but training them with reinforcement learning still depends on weak supervision: binary task success is too sparse for credit assignment, while frontier-langua → Open-source web agents are now strong enough to execute realistic browser tasks, but training them with reinforcement learning still depends on weak supervision: binary task success is too sparse for credit assignment, while frontier-langua
- 16 Oct 2026 Topics added cs.CL, cs.AI, cs.LG removed hf-daily-papers, conformal self-verification, compositional conformal certifier, polarity-aware lift, conformal trajectory selection, web agents, reinforcement learning, self-verifier
Changes detected by Silytics between refreshes of the source.
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.