Tev-Gym: 1,170 verified long-horizon agent environments with dense progress rewards
Wenbo Hu · hf · 2026-10-14
Adobe Research and UCLA introduce Tev, tackling the lack of data and sparse rewards in long-horizon agent training:
- Tev-Gym: the first long-horizon agent training dataset with verifiable, dense progressive rewards — 1,170 verified task environments across 7 domains, built latent-world-first from complete hidden task specifications
- PPRS: progress-potential reward shaping converts progress into dense per-step rewards, making intractable tasks trainable
- Tev-27B: matches GPT-6 Astra and Claude Opus 5 at Pass@5 on ALE computing/math domains, and generalizes to Terminal-Bench 3.0/4.0, surpassing GLM 5.2 and Gemini 3.7 Flash
More from coding & agent
- Codex autocompaction gets stuck: one steer and 10 million tokens later it's still parroting stale state — Aryvyo · 2026-10-14
- Agentic engineering vs AI engineering: Hugo Bowne-Anderson's clean split — Al_Grigor · 2026-10-14
- PostHog to devtools: stop asking 'how to add AI,' ask 'what's the agent-first version' — edwin · 2026-10-14
- Devin SWE-2 jumps to #3 on attention-to-detail coding test, tying Claude Sonnet 5.5 — silasalberti · 2026-10-14
- MCP hits ~500M monthly downloads but stays free — the money goes to key-management security firms — FinanceYF5 · 2026-10-14
- One-shot flying dragon: Gemini 4 Argon with Blender MCP and Tripo 3D builds a pure .blend scene — sidahuj · 2026-10-14