Pith Number

pith:LO2EULQN

pith:2022:LO2EULQNDEZ4R33CWWGIP7XQNV

not attested not anchored not stored refs resolved

Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?

Ari Holtzman, Hannaneh Hajishirzi, Luke Zettlemoyer, Mikel Artetxe, Mike Lewis, Sewon Min, Xinxi Lyu

Randomly replacing labels in in-context demonstrations barely hurts performance on classification and multiple-choice tasks across many models.

arxiv:2202.12837 v2 · 2022-02-25 · cs.CL · cs.AI

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{LO2EULQNDEZ4R33CWWGIP7XQNV}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

ground truth demonstrations are in fact not required -- randomly replacing labels in the demonstrations barely hurts performance on a range of classification and multi-choice tasks, consistently over 12 different models including GPT-3

C2weakest assumption

That randomly replacing labels does not introduce unintended statistical cues or that the chosen classification and multiple-choice tasks are representative of broader in-context learning behavior.

C3one line summary

Randomly replacing labels in in-context demonstrations barely hurts performance, showing that label space, input distribution, and sequence format drive in-context learning more than ground-truth labels.

References

237 extracted · 237 resolved · 6 Pith anchors

[1] Robust Disambiguation of Named Entities in Text 2011

[2] CODAH : An Adversarially-Authored Question Answering Dataset for Common Sense 2019

[3] Dolan, William B. and Brockett, Chris. Automatically Constructing a Corpus of Sentential Paraphrases. Proceedings of the Third International Workshop on Paraphrasing ( IWP 2005). 2005 2005

[4] DBpedia - A large-scale, multilingual knowledge base extracted from Wikipedia ,author=. Semantic Web ,year=

[5] Abductive Commonsense Reasoning ,author=. ICLR ,year=

Formal links

2 machine-checked theorem links

Cited by

32 papers in Pith

LLM4Log: A Systematic Review of Large Language Model-based Log Analysis

The Structure and Dynamics of the Online MAHA-sphere

Multilingual and Multimodal LLMs in the Wild: Building for Low-Resource Languages

OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences

Soft Head Selection for Injecting ICL-Derived Task Embeddings

Receipt and verification

First computed	2026-05-17T23:38:52.845449Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

5bb44a2e0d1933c8ef62b58c87fef06d4b52c2889253a1717819c66279b87b41

Aliases

arxiv: 2202.12837 · arxiv_version: 2202.12837v2 · doi: 10.48550/arxiv.2202.12837 · pith_short_12: LO2EULQNDEZ4 · pith_short_16: LO2EULQNDEZ4R33C · pith_short_8: LO2EULQN

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/LO2EULQNDEZ4R33CWWGIP7XQNV \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 5bb44a2e0d1933c8ef62b58c87fef06d4b52c2889253a1717819c66279b87b41

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "e4913c3c1646b32d0c2b76710aa070a3c523fa645f2f2b1702d4594d1213ece1",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2022-02-25T17:25:19Z",
    "title_canon_sha256": "274847321a77ccf35bc2559050cd585919998fa6fb270700a1bc1d6087907a26"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2202.12837",
    "kind": "arxiv",
    "version": 2
  }
}