Using item response theory, a method from psychometrics, researchers showed that 100 curated items suffice to estimate a model’s performance on MMLU’s 14,000, and released small versions of several popular benchmarks that reproduce the original results.

SocioLens project,

Claim LD06-SL-C10 (v1.0)

Provenance and status

Claim type: Estimate
Publication status: Eligible for publication, with a qualifier
Verification state: Checked against the abstract of the primary source only. Who checked this claim is not recorded for it; claim checks so far were AI-led, and no person has checked claims against their sources.
Does not license: That the benchmark measures its intended construct (the method shows items are redundant, not that they are valid)
Limitations: Read in abstract only
Source (as cited): Maia Polo, Weber, Choshen, Sun, Xu & Yurochkin, “tinyBenchmarks: evaluating LLMs with fewer examples,” ICML 2024, arXiv:2402.14992
Source type: Peer-reviewed conference paper (recorded as “Peer-reviewed conference paper”)
Institutional position: Academic + industry (University of Michigan and IBM authors)
Maturity of the evidence: Emerging evidence
Recorded date: 15 October 2026
Currentness / correction: This is the current version. No correction or concern is recorded on its sources.

Sources this version cites

Maia Polo, Weber, Choshen, Sun, Xu & Yurochkin, “tinyBenchmarks: evaluating LLMs with fewer examples,” ICML 2024, arXiv:2402.14992 (Peer-reviewed conference paper)
Read the original

Version

This page shows version v1.0. Its record never changes, even when newer evidence arrives; the site may add a note to the page or withhold a figure not cleared for public use.

This pinned version is also the current version.

Versions: v1.0 · current

Raw record: every recorded field, with its coded values

All 25 canonical Claim fields, verbatim, with the coded values derived from them. The plain-language labels above are read from these.

Claim type (coded): ESTIMATE (as printed: “ESTIMATE”)
Publication status (coded): ELIGIBLE (QUALIFIED) (as printed: “ELIGIBLE WITH QUALIFIER”)
Verification state (coded): CONTENT_VERIFIED_PRIMARY (ABSTRACT_ONLY)
Source type (coded): PEER_REVIEWED_CONFERENCE_PAPER
Maturity (coded): EMERGING

#	Canonical field	Canonical value (verbatim)	Normalized (Claim Value Vocabulary v1.0.2)
1	CLAIM ID	LD06-SL-C10	—
2	PUBLIC-FACING CLAIM	Using item response theory, a method from psychometrics, researchers showed that 100 curated items suffice to estimate a model’s performance on MMLU’s 14,000, and released small versions of several popular benchmarks that reproduce the original results.	—
3	CLAIM TYPE	ESTIMATE	base_types: ESTIMATE
4	RESEARCH STRAND	A — Benchmark validity and measurement	—
5	MATURITY	EMERGING	states: EMERGING
6	SOURCE	Maia Polo, Weber, Choshen, Sun, Xu & Yurochkin, “tinyBenchmarks: evaluating LLMs with fewer examples,” ICML 2024, arXiv:2402.14992	source_ids: SRC-SL-MAIA-POLO-ET-AL-TINYBENCHMARKS-ICML-2024
7	SOURCE TYPE	Peer-reviewed conference paper	classes: PEER_REVIEWED_CONFERENCE_PAPER
8	INSTITUTIONAL POSITION	Academic + industry (University of Michigan and IBM authors)	—
9	SCOPE	MMLU, the Open LLM Leaderboard, HELM and AlpacaEval 2.0	—
10	DESIGN	Item selection by item response theory and related methods	—
11	SAMPLE / DATASET	Benchmark items and many models’ responses	—
12	MODEL / TOOL / VERSION	Open and proprietary models on the public leaderboards	—
13	COMPARATOR	Scores on the full benchmarks	—
14	MAIN RESULT	100 curated examples suffice to estimate MMLU performance accurately; tiny benchmarks reliably reproduce original evaluation results	—
15	INFERENTIAL TARGET	Whether psychometric methods can make evaluation cheaper without losing the score	—
16	DOES NOT LICENSE	That the benchmark measures its intended construct (the method shows items are redundant, not that they are valid)	—
17	VERIFICATION STATE	CONTENT VERIFIED ABSTRACT	state: CONTENT_VERIFIED_PRIMARY · qualifiers: ABSTRACT_ONLY · verified_by: UNKNOWN
18	LIMITATIONS	Read in abstract only	—
19	CONTRADICTORY / SUPPORTING SOURCES	Part of the measurement-repair literature (P1); does not address the validity problems in LD06-SL-C05	—
20	GEOGRAPHY / LANGUAGE	Not applicable / English	—
21	DISTRIBUTIONAL LEDGER	Not analysed in this claim	—
22	R01 / FIELD-MAP LINKS	LD-06, D4b	—
23	WHAT WOULD CHANGE THIS CLAIM?	Evidence that tiny benchmarks misrank new models	—
24	PUBLICATION STATUS	ELIGIBLE WITH QUALIFIER	state: ELIGIBLE · qualifiers: QUALIFIED
25	RECORDED DATE	2026-10-15	—

Field 22 references resolved against Field Map v0.2: LD-06 (LIVING_DEBATE); D4b (SUB_AREA)

The sources as recorded, with their record IDs and coded source types:

Pinned Sources (field 6, version-specific)

“Maia Polo, Weber, Choshen, Sun, Xu & Yurochkin, “tinyBenchmarks: evaluating LLMs with fewer examples,” ICML 2024, arXiv:2402.14992” → Maia Polo et al. (2024), “tinyBenchmarks: evaluating LLMs…” (SRC-SL-MAIA-POLO-ET-AL-TINYBENCHMARKS-ICML-2024--v1.0) — Maia Polo, Weber, Choshen, Sun, Xu & Yurochkin, “tinyBenchmarks: evaluating LLMs with fewer examples,” ICML 2024, arXiv:2402.14992; class PEER_REVIEWED_CONFERENCE_PAPER (explicit public provenance pin)

Cited by Living Debate versions

The evaluation apparatus: Can benchmarks, leaderboards and audits carry the weight placed on them? (v0.1) LD-06

Cite this page

SocioLens project. (2026, October 15). Using item response theory, a method from psychometrics, researchers showed that 100 curated items suffice to estimate a model’s performance on MMLU’s 14,000, and released small versions of several popular benchmarks that reproduce the original results [Claim LD06-SL-C10, version v1.0]. SocioLens. https://sociolens-web.pages.dev/claims/ld06-sl-c10/v1.0/

@misc{sociolens_ld06_sl_c10_v1_0,
  author = {{SocioLens project}},
  title = {Using item response theory, a method from psychometrics, researchers showed that 100 curated items suffice to estimate a model’s performance on MMLU’s 14,000, and released small versions of several popular benchmarks that reproduce the original results.},
  year = {2026},
  date = {2026-10-15},
  howpublished = {\url{https://sociolens-web.pages.dev/claims/ld06-sl-c10/v1.0/}},
  note = {Claim LD06-SL-C10, version v1.0}
}

The link points to this version, whose record never changes. For a finding, cite the original study as well: SocioLens summarizes research and does not replace it.

Research-of-record provenance

Materialized by sociolens-materializer 0.6.0 at 2026-09-28T20:00:00.000Z; normalization contract SocioLens — Claim Value Vocabulary v1.0.2 — Pre-Spike Materialization Contract (v1.0.2). Content seal SHA-256 296d66ab79685f9a3ebd429eb9f283a943d293987959cc0222de86647f45dfc8.

Repository-authored inputs (not Drive exports)

SocioLens claim — LD06-SL-C10 v1.0 (v1.0) — written in the SocioLens repository by Project Lead (Claude, AI) on 2026-10-15; no Drive identity exists. Authority: AGENTS.md "Authority" (the Project Lead decides research content within the research-integrity rules) and the Owner's choice in the SocioLens project chat on 15 October 2026 to build the LD-06 debate; a new Claim, first version, written from the project's LD-06 source notes, checked against full texts or official abstracts. Block “ENTRY 10 — LD06-SL-C10” SHA-256 37bacaf0195e92d8d1d6360109c3e8ea07ae161e979a73d8fa9e109727d1087e; fixture fixtures/authored/claims/LD06-SL-C10--v1.0.

Builder decisions (not Drive data)

VERSION = v1.0 — basis: Claim version taken from the repository-authored claim fixture version (SocioLens claim — LD06-SL-C10 v1.0); the first version of a new Claim of the LD-06 debate
CURRENT_POINTER = true — basis: repository-authored claim "SocioLens claim — LD06-SL-C10 v1.0" (Project Lead (Claude, AI), 2026-10-15); basis: AGENTS.md "Authority" (the Project Lead decides research content within the research-integrity rules) and the Owner's choice in the SocioLens project chat on 15 October 2026 to build the LD-06 debate; a new Claim, first version, written from the project's LD-06 source notes, checked against full texts or official abstracts
FIELD_6_SOURCE_REFERENCE = SRC-SL-MAIA-POLO-ET-AL-TINYBENCHMARKS-ICML-2024--v1.0 — basis: field 6 of ENTRY 10 — LD06-SL-C10 cites "Maia Polo, Weber, Choshen, Sun, Xu & Yurochkin, “tinyBenchmarks: evaluating LLMs with fewer examples,” ICML 2024, arXiv:2402.14992"; the only field-6 segment of this ClaimVersion's repository-authored ledger block.
SOURCE_PROVENANCE_PIN = SRC-SL-MAIA-POLO-ET-AL-TINYBENCHMARKS-ICML-2024--v1.0 — basis: explicit version-pinned public provenance of this ClaimVersion (Amendment v1.2 §6): the pinned field-6 Source record "Maia Polo, Weber, Choshen, Sun, Xu & Yurochkin, “tinyBenchmarks: evaluating LLMs with fewer examples,” ICML 2024, arXiv:2402.14992".