AI Paper Atlas

Research map / Large language models / AI and society

BRIDGE: Predicting Human Task Completion Time From Model Performance

Fengyuan Liu, Jay Gala, Nilaksh, Dzmitry Bahdanau, Siva Reddy, Hugo Larochelle

ICML 2026 · Large language models › AI and society · approach: Social-science view of AI evaluation · 0 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use IRT latent difficulty instead of logit success-rate difficulty to predict human timeBRIDGE · CybenchR2: -1.366 -> 0.450Improved
use IRT difficulty instead of LLM-predicted time (Gemini 3 Pro, GPT-5.2)BRIDGE · Cybenchwithin 2x tolerance: 92% vs 38% and 8%Improved
use IRT difficulty instead of logit success rateBRIDGE · SWE-bench Verifiedbucket accuracy: 38.6% -> 41.6%Improved
model task discrimination and latent ability with 2PL IRT rather than only success ratesBRIDGE · SWE-bench Verifiedweighted kappa: 0.115 -> 0.231Improved
randomly remove up to 50% of observed model-task responses (ablation)BRIDGE IRT fit · METR plus OOD benchmarksSpearman rho(b): 0.9947 at 10% removed -> 0.9735 at 50% removedNo clear effect
remove 70% of observed responses (ablation)BRIDGE IRT fit · METR plus OOD benchmarksSpearman rho(b): 0.9424Hurt
raise success threshold from 50% to 80% (ablation)BRIDGE forecast · frontier task horizonsolvable task length: from ~1-2.5 hours to mostly under 1 hourHurt
raise success threshold from 50% to 80% (ablation)BRIDGE forecast · horizon doubling timedoubling timeNo clear effect
calibrate latent difficulty log-linearly to METR human time2PL IRT difficulty · METR tasksR2: R2 = 0.81; 2.26x time per +1 in bImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy