Pith. sign in

REVIEW 4 cited by

Measuring Causal Effects of Data Statistics on Language Model's `Factual' Predictions

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 2207.14251 v2 pith:TTIOGXLW submitted 2022-07-28 cs.CL

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

Large amounts of training data are one of the major reasons for the high performance of state-of-the-art NLP models. But what exactly in the training data causes a model to make a certain prediction? We seek to answer this question by providing a language for describing how training data influences predictions, through a causal framework. Importantly, our framework bypasses the need to retrain expensive models and allows us to estimate causal effects based on observational data alone. Addressing the problem of extracting factual knowledge from pretrained language models (PLMs), we focus on simple data statistics such as co-occurrence counts and show that these statistics do influence the predictions of PLMs, suggesting that such models rely on shallow heuristics. Our causal framework and our results demonstrate the importance of studying datasets and the benefits of causality for understanding NLP models.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

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

  1. When LLMs Stop Following Steps: A Diagnostic Study of Procedural Execution in Language Models

    cs.CL 2026-05 unverdicted novelty 7.0 of 10

    LLM first-answer accuracy on procedural arithmetic drops from 61% on 5-step tasks to 20% on 95-step tasks, with frequent failures including skipped steps, premature answers, and hallucinated operations.

  2. Tracing LLM Behavior to the Training Data with Empirical Next-Token Distributions

    cs.AI 2026-07 conditional novelty 6.0 of 10

    A trained transformer's next-token distribution often matches the empirical next-token distribution of its pretraining corpus, with agreement improving as models grow, while a persistent tail of mismatches remains.

  3. The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text

    cs.CL 2025-06 conditional novelty 6.0 of 10

    A new 8TB openly-licensed text corpus trains 7B LLMs that are competitive with Llama 1/2, showing that performant models need not depend on unlicensed web data.

  4. Protoknowledge Shapes Behaviour of LLMs in Downstream Tasks: Memorization and Generalization with Knowledge Graphs

    cs.CL 2025-05 conditional novelty 5.0 of 10

    A new 'protoknowledge' analysis links lexical, hierarchical, and topological knowledge graph recall in LLMs to Text-to-SPARQL performance, offering a black-box route to detect semantic contamination.

Pith tools