Ai2's StateBench Finds Frontier Models Failing at Long State Tracking

Ai2 introduced StateBench, a benchmark with four complexity tiers for state tracking, on which frontier LLMs score as low as 13% on long tasks while recurrent architectures perform better.

Read the full story at AlphaSignal ↗

Timeline · 1 report

2026-11-03 17:28 · AlphaSignal
Ai2's StateBench Catches Frontier Models Failing at 13% on Long State Tracking

More stories

Postman's silent self-install into Claude Code and Codex sparks security concerns — Runway Research
Democrats' midterm wins cloud outlook for AI data center boom — DigiTimes Asia
Google to cut off search access for schools that ban AI, drawing political backlash — Forbes AI
UK cyber agency warns AI-powered malware likely to hit critical infrastructure by 2028 — The Guardian AI
OpenAI weighs rebranding data centers as 'AI compute campuses' amid growing public backlash — Techmeme
Anthropic reports 1,100% revenue surge and $895B compute commitments ahead of IPO — Reuters AI
AI inference startup Fal raises $1.3B at $16B valuation in Bessemer-led round — Bloomberg AI
OpenAI hires Uber President Jill Hazelbaker as chief corporate affairs officer — OpenAI News

Get the daily brief of stories like this at 6:30 every morning →