Pith. sign in

REVIEW 2 cited by

Revisiting Out-of-distribution Robustness in NLP: Benchmark, Analysis, and LLMs Evaluations

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 2306.04618 v2 pith:KGXYUFTU submitted 2023-06-07 cs.CL cs.CRcs.LG

Revisiting Out-of-distribution Robustness in NLP: Benchmark, Analysis, and LLMs Evaluations

classification cs.CL cs.CRcs.LG
keywords robustnessllmsbenchmarkbossevaluationfindfine-tuningmodels
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

This paper reexamines the research on out-of-distribution (OOD) robustness in the field of NLP. We find that the distribution shift settings in previous studies commonly lack adequate challenges, hindering the accurate evaluation of OOD robustness. To address these issues, we propose a benchmark construction protocol that ensures clear differentiation and challenging distribution shifts. Then we introduce BOSS, a Benchmark suite for Out-of-distribution robustneSS evaluation covering 5 tasks and 20 datasets. Based on BOSS, we conduct a series of experiments on pre-trained language models for analysis and evaluation of OOD robustness. First, for vanilla fine-tuning, we examine the relationship between in-distribution (ID) and OOD performance. We identify three typical types that unveil the inner learning mechanism, which could potentially facilitate the forecasting of OOD robustness, correlating with the advancements on ID datasets. Then, we evaluate 5 classic methods on BOSS and find that, despite exhibiting some effectiveness in specific cases, they do not offer significant improvement compared to vanilla fine-tuning. Further, we evaluate 5 LLMs with various adaptation paradigms and find that when sufficient ID data is available, fine-tuning domain-specific models outperform LLMs on ID examples significantly. However, in the case of OOD instances, prioritizing LLMs with in-context learning yields better results. We identify that both fine-tuned small models and LLMs face challenges in effectively addressing downstream tasks. The code is public at \url{https://github.com/lifan-yuan/OOD_NLP}.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

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

  1. Testing LLM Arithmetic Reasoning Generalization with Automatic Numeric-Remapping Attacks

    cs.CR 2026-06 unverdicted novelty 7.0

    An automatic numeric-remapping attack generator reveals 12-26 point accuracy drops on GSM8K for three LLMs while MAWPS and MultiArith stay near 98%.

  2. STELLAR-E: a Synthetic, Tailored, End-to-end LLM Application Rigorous Evaluator

    cs.AI 2026-04 unverdicted novelty 5.0

    STELLAR-E modifies the TGRT Self-Instruct framework to produce tailored synthetic LLM evaluation datasets that score an average 5.7% higher on LLM-as-a-judge metrics than existing language-specific benchmarks.