MétaCan
Menu
Back to cohort

Systematic Generalization by Finetuning? Analyzing Pretrained Language Models Using Constituency Tests

2023· article· en· W4389518176 on OpenAlexafffund
Aishik Chakraborty, Jackie CK Cheung, Timothy J. O’Donnell

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsMcGill University
FundersSamsungCompute Canada
KeywordsSentenceComputer scienceNatural language processingArtificial intelligenceGeneralizationCopyingReplicateLanguage modelWord (group theory)Linguistics

Abstract

fetched live from OpenAlex

Constituents are groups of words that behave as a syntactic unit.Many linguistic phenomena (e.g., question formation, diathesis alternations) require the manipulation and rearrangement of constituents in a sentence.In this paper, we investigate how different finetuning setups affect the ability of pretrained sequence-to-sequence language models such as BART and T5 to replicate constituency tests -transformations that involve manipulating constituents in a sentence.We design multiple evaluation settings by varying the combinations of constituency tests and sentence types that a model is exposed to during finetuning.We show that models can replicate a linguistic transformation on a specific type of sentence that they saw during finetuning, but performance degrades substantially in other settings, showing a lack of systematic generalization.These results suggest that models often learn to manipulate sentences at a surface level unrelated to the constituent-level syntactic structure, for example by copying the first word of a sentence.These results may partially explain the brittleness of pretrained language models in downstream tasks 1 .

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.005
metaresearch head score (Gemma)0.032
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.005
Threshold uncertainty score0.025

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0050.032
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0000.000
Science and technology studies0.0000.001
Scholarly communication0.0010.004
Open science0.0010.001
Research integrity0.0010.003
Insufficient payload (model declined to judge)0.0020.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.033
GPT teacher head0.278
Teacher spread0.244 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2023
Admission routes2
Has abstractyes

Explore more

Same topicTopic ModelingFrench-language works237,207