Pith. sign in

REVIEW 11 cited by

Are Emergent Abilities of Large Language Models a Mirage?

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 2304.15004 v2 pith:4GEIPNFJ submitted 2023-04-28 cs.AI cs.LG

classification cs.AIcs.LG
keywords abilitiesemergentmodelmodelsmetricsmetricproduceseemingly
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Recent work claims that large language models display emergent abilities, abilities not present in smaller-scale models that are present in larger-scale models. What makes emergent abilities intriguing is two-fold: their sharpness, transitioning seemingly instantaneously from not present to present, and their unpredictability, appearing at seemingly unforeseeable model scales. Here, we present an alternative explanation for emergent abilities: that for a particular task and model family, when analyzing fixed model outputs, emergent abilities appear due to the researcher's choice of metric rather than due to fundamental changes in model behavior with scale. Specifically, nonlinear or discontinuous metrics produce apparent emergent abilities, whereas linear or continuous metrics produce smooth, continuous predictable changes in model performance. We present our alternative explanation in a simple mathematical model, then test it in three complementary ways: we (1) make, test and confirm three predictions on the effect of metric choice using the InstructGPT/GPT-3 family on tasks with claimed emergent abilities; (2) make, test and confirm two predictions about metric choices in a meta-analysis of emergent abilities on BIG-Bench; and (3) show to choose metrics to produce never-before-seen seemingly emergent abilities in multiple vision tasks across diverse deep networks. Via all three analyses, we provide evidence that alleged emergent abilities evaporate with different metrics or with better statistics, and may not be a fundamental property of scaling AI models.

Discussion (0). Sign in to comment.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 133 citations worldwide. Full citation record

  1. Context Is King: How In-Context Specification Shapes the Geometry of Concepts

    cs.LG 2026-07 accept novelty 7.5 of 10

    In capable Gemma and Qwen models, declarative in-context rules set the relational geometry and topology type that the model represents and causally uses, overriding strong pretrained priors.

  2. Training with (Swap) Regret Loss in a Single-Layer Self-Attention Model: A Case Study on the Probability Simplex

    cs.LG 2026-07 conditional novelty 7.0 of 10

    Training single-layer attention with squared regret loss has stationary points that implement smoothed fictitious play (external regret) and, via a new swap-regret loss, the Blum–Mansour no-swap-regret algorithm.

  3. Transplanting, inverting, and preventing a misalignment persona: method-conditional emergent misalignment in Qwen2.5

    cs.CL 2026-07 conditional novelty 7.0 of 10

    Emergent misalignment in Qwen2.5 is mediated by a causal persona direction that low-rank LoRA recruits from covert code while full SFT does not and moves against it.

  4. Will Scaling Improve Social Simulation with LLMs?

    cs.CL 2026-07 conditional novelty 7.0 of 10

    Using 85 controlled and 35 public LLMs, the authors show social-simulation accuracy generally improves with compute, but some behavioral and low-resource tasks do not scale.

  5. FactorJEPA: Factorizing Monolithic Futures into Layout-Agent-Interaction Channels for Crowded and Chaotic Global South Urban Worlds

    cs.AI 2026-08 conditional novelty 6.0 of 10

    FactorJEPA splits a video prediction model into layout, agent, and interaction channels with a visibility gate, and a new DENSEWORLD dataset tests it on crowded Indian city scenes.

  6. Grokking Is Conditional and Fragile: A Fully-Tractable, Multi-Seed Study at 12K Parameters

    cs.LG 2026-07 accept novelty 6.0 of 10

    In a fully tractable 12K Llama-style model, grokking is a conditional fragile phase transition gated by coverage (tracking modulus more than structure), weight decay, and floating-point reduction order, so evidence mu...

  7. RLearner-LLM: Balancing Logical Grounding and Fluency in Large Language Models via Hybrid Direct Preference Optimization

    cs.CL 2026-05 unverdicted novelty 6.0 of 10

    RLearner-LLM's Hybrid-DPO fuses DeBERTa NLI and LLM verifier scores to deliver up to 6x higher NLI entailment than standard SFT while preserving answer coverage across academic domains.

  8. Artificial or Human Intelligence?

    econ.TH 2025-09 conditional novelty 6.0 of 10

    A theoretical model shows that AI's sharp capability cutoff and hallucination rate create a discontinuous gap in student ability around the AI frontier, and that AI-free assignments can correct students' overestimatio...

  9. What Does it Mean for a Neural Network to Learn a "World Model"?

    cs.AI 2025-07 conditional novelty 6.0 of 10

    Defines a world model as a simple commutative-diagram factorization through an intermediate representation, with conditions that the model be learned and emergent rather than inherited from input or output.

  10. Bottom-up Domain-specific Superintelligence: A Reliable Knowledge Graph is What We Need

    cs.CL 2025-07 conditional novelty 6.0 of 10

    A language model fine-tuned on knowledge-graph-path reasoning tasks (QwQ-Med-3) beats strong baselines on a same-style benchmark but shows mixed gains on external medical QA tests.

  11. Foundation Models for Astrophysics

    astro-ph.IM 2026-08 conditional novelty 3.0 of 10

    Astronomical 'foundation models' largely reuse transformers and self-supervised pretraining, but evidence of transfer to new instruments, populations, or tasks remains rare; the paper argues such evidence, not archite...

Pith tools