ByteDance Paper: Keeping Full History Cuts Costs by 19% vs. Compression
AGI Hunt · wechat · 2026-08-31
ByteDance's Chain-of-Experience paper investigates how AI agents should handle past attempts during iterative solving. Contrary to intuition that compression saves context, experiments show retaining the full history with feedback signals achieves a 71.0% average accuracy across 8 models and 6 benchmarks, outperforming non-iterative solving (66.8%). Despite longer context windows, faster convergence reduces total API costs by 19%. The study suggests that feedback on "why it went wrong" holds more value than compressed result summaries, validating a "less is more" context strategy.
More from coding & agent
- Developer Critiques MCP Spec: Overcomplicated JSON-RPC? — PaulMorel · 2026-09-01
- VibeKit MCP Server Manages Deployments, Logs, and Headless Coding — modelcontextprotocol · 2026-09-01
- Distributed.systems Launches Auditable Agent Infrastructure — arthurcolle · 2026-09-01
- How to Stop Context Window Bottlenecks in Data-Heavy MCP Servers — JuicerSocial · 2026-09-01
- Grok Bots Turn LLM Citations Into an SEO Loop for AI Search Ranking — rohanpaul_ai · 2026-09-01
- Search configuration impacts agent accuracy 40x more than model choice — RichardSocher · 2026-09-01