NVIDIA and Stanford's CLM Treats Decision-Making as Retrieval, Up to 9x Faster
wschroll · x · 2026-09-26
NVIDIA and Stanford's Contrastive Language Model (CLM) skips text generation and handles small repeated decisions inside AI systems: choosing tools, ranking patches, routing requests, picking next actions. It treats decision-making as retrieval — encoding state with a frozen Qwen3-8B, then contrastively retrieving decisions — claiming up to 9x faster System 1 decision-making. The thread also catalogs the Jev ecosystem: live trading bots, autonomous browser agents, frame-by-frame Doom agents.
More from coding & agent
- Pigment Agents now reads and posts in Slack, inheriting user permissions — mattturck · 2026-11-12
- Map of the agent economy: 50+ startups across payments, spend control, identity, memory — venturetwins · 2026-11-12
- OpenCode launches Browse, an on-device browser agent with local logins — Saul_Loveman · 2026-11-12
- VentureTwins maps the B2A economy: dozens of agent infra and app companies in first series installment — venturetwins · 2026-11-12
- Google releases 76-page 'Agents Companion' whitepaper on agent engineering — mdancho84 · 2026-11-12
- Spinning up containers for 20-line agent tools is backwards: Safescript compiles provable, 50ms-deploy apps — uriwa · 2026-11-12