Pith. sign in

REVIEW 17 cited by

Dynabench: Rethinking Benchmarking in NLP

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2104.14337 v1 pith:5JAIX3ZM submitted 2021-04-07 cs.CL cs.AI

Dynabench: Rethinking Benchmarking in NLP

classification cs.CL cs.AI
keywords dynabenchmodelbenchmarkingcreationdatasetdynamicexamplesplatform
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We introduce Dynabench, an open-source platform for dynamic dataset creation and model benchmarking. Dynabench runs in a web browser and supports human-and-model-in-the-loop dataset creation: annotators seek to create examples that a target model will misclassify, but that another person will not. In this paper, we argue that Dynabench addresses a critical need in our community: contemporary models quickly achieve outstanding performance on benchmark tasks but nonetheless fail on simple challenge examples and falter in real-world scenarios. With Dynabench, dataset creation, model development, and model assessment can directly inform each other, leading to more robust and informative benchmarks. We report on four initial NLP tasks, illustrating these concepts and highlighting the promise of the platform, and address potential objections to dynamic benchmarking as a new standard for the field.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 17 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Meta-Benchmarks for Financial-Services LLM Evaluation

    cs.AI 2026-07 unverdicted novelty 7.0

    A meta-benchmarking framework organizes 452 LLM benchmarks into 41 O*NET Generalized Work Activities and 38 BIAN domains, using discrimination-coverage-recency weights to scale K-factors in an Elo tournament for compa...

  2. CoEval: Ranking Language Models for Custom Tasks Without Labeled Data or Trustworthy Benchmarks

    cs.CL 2026-06 conditional novelty 7.0

    CoEval generates task-specific benchmarks by rotating models through teacher, student, and judge roles, then weights questions by discriminative power and judges by panel consensus to recover accurate model rankings w...

  3. Agent Island: A Saturation- and Contamination-Resistant Benchmark from Multiagent Games

    cs.AI 2026-05 unverdicted novelty 7.0

    Agent Island is a new multiagent game environment that functions as a dynamic benchmark resistant to saturation and contamination, with Bayesian ranking showing OpenAI GPT-5.5 as the strongest performer among 49 model...

  4. TRIP-Evaluate: An Open Multimodal Benchmark for Evaluating Large Models in Transportation

    cs.CV 2026-04 accept novelty 7.0

    TRIP-Evaluate is a new open multimodal benchmark with 837 text, image, and point-cloud items organized by a role-task-knowledge taxonomy to evaluate large models on transportation workflows.

  5. Analysis and Explainability of LLMs Via Evolutionary Methods

    cs.NE 2026-04 unverdicted novelty 7.0

    Evolutionary trees from LLM weights recover ground-truth training topologies and identify key datasets and layers through phenotypic analysis.

  6. CT Open: An Open-Access, Uncontaminated, Live Platform for the Open Challenge of Clinical Trial Outcome Prediction

    cs.AI 2026-04 accept novelty 7.0

    CT Open is a new live platform with an automated LLM-powered decontamination pipeline that supplies uncontaminated benchmarks for predicting clinical trial outcomes.

  7. Open-World Evaluations for Measuring Frontier AI Capabilities

    cs.AI 2026-05 conditional novelty 6.0

    Open-world evaluations using qualitative review of real-world tasks can give earlier warnings of frontier AI capabilities than automated benchmarks, as demonstrated by an AI agent publishing a simple iOS app with one ...

  8. Navigating the Sea of LLM Evaluation: Investigating Bias in Toxicity Benchmarks

    cs.AI 2026-05 unverdicted novelty 6.0

    Toxicity benchmarks for LLMs produce inconsistent results when task type, input domain, or model changes, revealing intrinsic evaluation biases.

  9. QuickScope: Certifying Hard Questions in Dynamic LLM Benchmarks

    cs.CL 2026-04 unverdicted novelty 6.0

    QuickScope uses modified COUP Bayesian optimization to find truly difficult questions in dynamic LLM benchmarks more sample-efficiently than baselines while cutting false positives.

  10. RoboPlayground: Democratizing Robotic Evaluation through Structured Physical Domains

    cs.RO 2026-04 unverdicted novelty 6.0

    RoboPlayground reframes robotic manipulation evaluation as a language-driven process over structured physical domains, letting users author varied yet reproducible tasks that reveal policy generalization failures.

  11. Beyond Benchmark Islands: Toward Representative Trustworthiness Evaluation for Agentic AI

    cs.CL 2026-03 unverdicted novelty 6.0

    Defines agentic trustworthiness via five properties and proposes HAAF, a scenario-distribution framework with a Trustworthy Optimization Factory that transfers interventions across 13 models from seven families on a 1...

  12. Inflated Excellence or True Performance? Rethinking Medical Diagnostic Benchmarks with Dynamic Evaluation

    cs.CL 2025-10 conditional novelty 6.0

    DyReMe generates fresh, confounded medical consultation cases to expose weaknesses in state-of-the-art LLMs across accuracy, veracity, helpfulness, and consistency beyond static benchmarks.

  13. League of LLMs: A Benchmark-Free Paradigm for Mutual Evaluation of Large Language Models

    cs.AI 2025-07 unverdicted novelty 6.0

    League of LLMs organizes LLMs into a self-governed mutual evaluation league using dynamic, transparent, objective, and professional criteria to distinguish model capabilities with 70.7% top-k ranking stability.

  14. Interactive Evaluation Requires a Design Science

    cs.AI 2026-05 unverdicted novelty 5.0

    Interactive evaluation of AI must be reframed as a distinct paradigm that maps interaction trajectories to judgments on process, recoverability, coordination, robustness, and system performance, supported by a two-axi...

  15. Too long; didn't solve

    cs.AI 2026-04 unverdicted novelty 5.0

    Longer prompts and solutions in a new expert-authored math dataset correlate with higher failure rates across LLMs, with length linked to empirical difficulty after difficulty adjustment.

  16. Humanity's Last Exam

    cs.LG 2025-01 unverdicted novelty 5.0

    Humanity's Last Exam is a new 2,500-question benchmark at the frontier of human knowledge where state-of-the-art LLMs show low accuracy.

  17. Too long; didn't solve

    cs.AI 2026-04 conditional novelty 4.0

    Prompt length and solution length both rise with LLM failure on expert-authored adversarial math problems, linking structural length to empirical difficulty.