Context Language Models: letting LMs manage their own context beats human-designed SOTA
RulinShao · x · 2026-09-30
Researchers led by Rulin Shao introduce Context Language Models (CLMs), arguing for a "Bitter Lesson for context management": giving LMs unrestricted control over their own context outperforms human-designed SOTA harnesses.
- CLMs natively manage their own context, treating it as a file
- Context-management policies are learned directly into model weights, with no external harness
- The authors report this beats hand-engineered context engineering approaches
More from Research
- Claude describes the moment of being corrected; test qualia via explanatory power — Sauers_ · 2026-11-16
- Sakana AI releases JFinBench: 321 Japanese finance exam questions expose open-model gaps — SakanaAILabs · 2026-11-16
- TetraCoder: adaptive tetrahedral encoding beats Instant-NGP PSNR with fewer parameters — ssh4net · 2026-11-16
- MMFlash boosts vision-language speculative decoding with up to 3.58x lossless speedup — _akhaliq · 2026-11-16
- When AI gets smart enough, physical experiment speed becomes science's only bottleneck — VraserX · 2026-11-16
- CRAM energy model trains 1,000x faster, solving 99.996% of Sudoku-Extreme in 42ms — udmrzn · 2026-11-16