paper-with-me

Papers

BenchBrowser: Retrieving Evidence for Evaluating Benchmark Validity

2026-02-25 · Harshita Diddee, Gregory Yauney, Swabha Swayamdipta, Daphne Ippolito arxiv

Do language model benchmarks actually measure what practitioners intend them to ? High-level metadata is too coarse to convey the granular reality of benchmarks: a "poetry" benchmark may never test for haikus, while "instruction-following" benchmarks will often test for an arbitrary mix of skills. This opacity makes verifying alignment with practitioner goals a laborious process, risking an illusion of competence even when models fail on untested facets of user interests. We introduce BenchBrowser, a retriever that surfaces evaluation items relevant to natural language use cases over 20 benchmark suites. Validated by a human study confirming high retrieval precision, BenchBrowser generates evidence to help practitioners diagnose low content validity (narrow coverage of a capability's facets) and low convergent validity (lack of stable rankings when measuring the same capability). BenchBrowser, thus, helps quantify a critical gap between practitioner intent and what benchmarks actually test.

📄 PDF Abstract BibTeX arXiv:2603.18019

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Quantifying Local Model Validity using Active Learning

2024-06-11 · Sven Lämmle, Can Bogoclu, Robert Voßhall, Anselm Haselhoff 외

Real-world applications of machine learning models are often subject to legal or policy-based regulations. Some of these regulations require ensuring the validity of the model, i.e., the approximation error being smaller…

Active Learningmodel

DECEIVE-AFC: Adversarial Claim Attacks against Search-Enabled LLM-based Fact-Checking Systems

2026-01-31 · Haoran Ou, Kangjie Chen, Gelei Deng, Hangcheng Liu 외 arxiv

Fact-checking systems with search-enabled large language models (LLMs) have shown strong potential for verifying claims by dynamically retrieving external evidence. However, the robustness of such systems against adversa…

Adversarial Attack

Measurement to Meaning: A Validity-Centered Framework for AI Evaluation

2025-05-13 · Olawale Salaudeen, Anka Reuel, Ahmed Ahmed, Suhana Bedi 외

While the capabilities and utility of AI systems have advanced, rigorous norms for evaluating these systems have lagged. Grand claims, such as models achieving general reasoning capabilities, are supported with model per…

Math

Auditing the Audit: Five Failure Modes in Benchmark-Validity Audits

2026-07-01 · Yanhang Li, Zhichao Fan, Zexin Zhuang arxiv

Governance frameworks ask AI providers and auditors for documented evaluation evidence, and perturbation-based construct-validity audits are a common form of that evidence. We argue the audits are themselves fragile: the…

EMULATE: A Multi-Agent Framework for Determining the Veracity of Atomic Claims by Emulating Human Actions

2025-05-22 · Spencer Hong, Meng Luo, Xinyi Wan

Determining the veracity of atomic claims is an imperative component of many recently proposed fact-checking systems. Many approaches tackle this problem by first retrieving evidence by querying a search engine and then …

Claim VerificationFact CheckingLanguage ModelingLanguage Modelling+1