REVIEW 12 cited by
Impact of Pretraining Term Frequencies on Few-Shot Reasoning
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
Impact of Pretraining Term Frequencies on Few-Shot Reasoning
read the original abstract
Pretrained Language Models (LMs) have demonstrated ability to perform numerical reasoning by extrapolating from a few examples in few-shot settings. However, the extent to which this extrapolation relies on robust reasoning is unclear. In this paper, we investigate how well these models reason with terms that are less frequent in the pretraining data. In particular, we examine the correlations between the model performance on test instances and the frequency of terms from those instances in the pretraining data. We measure the strength of this correlation for a number of GPT-based language models (pretrained on the Pile dataset) on various numerical deduction tasks (e.g., arithmetic and unit conversion). Our results consistently demonstrate that models are more accurate on instances whose terms are more prevalent, in some cases above $70\%$ (absolute) more accurate on the top 10\% frequent terms in comparison to the bottom 10\%. Overall, although LMs exhibit strong performance at few-shot numerical reasoning tasks, our results raise the question of how much models actually generalize beyond pretraining data, and we encourage researchers to take the pretraining data into account when interpreting evaluation results.
Forward citations
Cited by 12 Pith papers
-
Gradient-Based Program Synthesis with Neurally Interpreted Languages
NLI autonomously discovers a vocabulary of primitive operations and interprets variable-length programs via a neural executor, allowing end-to-end training and gradient-based test-time adaptation that outperforms prio...
-
Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling
Pythia releases 16 identically trained LLMs with full checkpoints and data tools to study training dynamics, scaling, memorization, and bias in language models.
-
Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?
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.
-
The Right Answer, the Wrong Direction: Why Transformers Fail at Counting and How to Fix It
Transformers store count information internally but cannot read it out as digits due to near-orthogonal alignment with output-head rows; updating digit rows or applying LoRA to attention layers improves constrained an...
-
The Right Answer, the Wrong Direction: Why Transformers Fail at Counting and How to Fix It
Transformers encode counts correctly internally but fail to read them out due to misalignment with digit output directions, fixable by updating 37k output parameters or small LoRA on attention.
-
LLM-as-a-Coach: Experiential Learning for Non-Verifiable Tasks
Training a language model by distilling a coach's written experiential knowledge beats training on a scalar rubric score for open-ended tasks, with better out-of-distribution transfer.
-
FoNE: Precise Single-Token Number Embeddings via Fourier Features
FoNE encodes numbers as single tokens via Fourier features and outperforms subword and digit-wise embeddings on addition, subtraction, and multiplication with far less data.
-
Lessons from the Trenches on Reproducible Evaluation of Language Models
The paper compiles practical lessons on reproducible LM evaluation and introduces the lm-eval library to mitigate common methodological problems in NLP.
-
Emergent Abilities of Large Language Models
Emergent abilities are capabilities present in large language models but absent in smaller ones and cannot be predicted by extrapolating smaller model performance.
-
GPT-NeoX-20B: An Open-Source Autoregressive Language Model
GPT-NeoX-20B is a publicly released 20B parameter autoregressive language model trained on the Pile that shows strong gains in five-shot reasoning over similarly sized prior models.
-
Lost in Cultural Translation: Do LLMs Struggle with Math Across Cultural Contexts?
LLMs show accuracy drops of 0.3% to 5.9% on GSM8K math problems when culturally adapted to six countries while keeping math operations identical, with statistical significance confirmed by McNemar tests.
-
Galactica: A Large Language Model for Science
Galactica, a science-specialized LLM, reports higher scores than GPT-3, Chinchilla, and PaLM on LaTeX knowledge, mathematical reasoning, and medical QA benchmarks while outperforming general models on BIG-bench.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.