NeurIPS 2021 re-review flipped 23% of decisions; only ~1/3 of claims overlap
JessicaHullman · x · 2026-10-30
Quantifying the lottery variance of peer review: the NeurIPS 2021 experiment saw 23% of papers flip decisions on re-review. A deeper claim-level analysis finds only 1/3 of claims appear across the committee, showing coverage is highly fragmented. Reviewers also err — among directly checkable claims, the paper refutes the criticism 18% of the time.
More from Research
- Commonroom beta: chat with neuron-steered LLM personas and hunt emergent personalities — repligate · 2026-10-30
- Elorian launches visual reasoning model Glimpse alongside DecoyCount benchmark — AndrewDai · 2026-10-30
- Instruction Capacity paper quantifies rule-following limits: GPT-4o mini holds ~37 rules at 95% accuracy — Hennadii Kutomanov · 2026-10-30
- Guidelabs launches Causal Diffusion Engine, adding interpretability to diffusion LLMs with no latency cost — juliusadml · 2026-10-30
- Alice & Bob says RSA-2048 falls in 3.14 days with 23,000 cat qubits — jedisct1 · 2026-10-30
- FlexGlue's 0.6 px average keypoint shift matches SIFT's subpixel refinement threshold — ducha_aiki · 2026-10-30