Dispute: Claiming 100% on benchmark without private eval is invalid
JFPuget · x · 2026-08-31
JFPuget responded to criticism about their model's score on the ARC benchmark, clarifying they only tested on public evaluation games and noting the lack of tools for semi-private data evaluation. François Chollet countered that claiming a score on a benchmark without private evaluation is invalid and criticized the decision not to open source the solution.
Related event: Coding Agent's ARC-AGI-3 100% Claim Rebutted by Chollet(6 posts)→
More from Models
- Why are sub-15B open models so rare now? Labs seem to have settled on 24-27B minimum — Hot_Example_4456 · 2026-10-16
- User spots model openly generating "ideal responses" to train other models — liminal_bardo · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- Gemma 4 31B QAT on one 24GB card: 46 tok/s plain, ~80 with draft model on RTX 4090 — Subject_Poet_3861 · 2026-10-16
- JevBench v1.7.46 leaderboard: new d1-ultra tops 32 API systems at 76.1 composite — airesearch12 · 2026-10-16
- Team of 5 mulls dropping ChatGPT Pro for Business as OpenAI quota parity stays unclear — bettercallfool · 2026-10-16