REVIEW 15 cited by
A Careful Examination of Large Language Model Performance on Grade School Arithmetic
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
read the original abstract
Large language models (LLMs) have achieved impressive success on many benchmarks for mathematical reasoning. However, there is growing concern that some of this performance actually reflects dataset contamination, where data closely resembling benchmark questions leaks into the training data, instead of true reasoning ability. To investigate this claim rigorously, we commission Grade School Math 1000 (GSM1k). GSM1k is designed to mirror the style and complexity of the established GSM8k benchmark, the gold standard for measuring elementary mathematical reasoning. We ensure that the two benchmarks are comparable across important metrics such as human solve rates, number of steps in solution, answer magnitude, and more. When evaluating leading open- and closed-source LLMs on GSM1k, we observe accuracy drops of up to 8%, with several families of models showing evidence of systematic overfitting across almost all model sizes. Further analysis suggests a positive relationship (Spearman's r^2 = 0.36) between a model's probability of generating an example from GSM8k and its performance gap between GSM8k and GSM1k, suggesting that some models may have partially memorized GSM8k. Nevertheless, many models, especially those on the frontier, show minimal signs of overfitting, and all models broadly demonstrate generalization to novel math problems guaranteed to not be in their training data.
Forward citations
Cited by 15 Pith papers
-
RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents
RigorBench is the first benchmark for process discipline in autonomous AI coding agents, reporting 41% higher process quality scores and 17% higher outcome correctness when agents follow structured engineering practices.
-
LiveBench: A Challenging, Contamination-Limited LLM Benchmark
LiveBench is a contamination-limited LLM benchmark with auto-scored challenging tasks from recent sources across math, coding, reasoning and more, where top models score below 70%.
-
TLA+-Bench: An Execution-Grounded Benchmark and Dataset for Natural-Language to TLA Specification Generation
An exact model-checker oracle for NL-to-TLA+ still yields an elevenfold “correctness envelope” (18.7%–1.7%) once interface supply and vacuity screens are made explicit.
-
RigorBench: Benchmarking Engineering Process Discipline in Autonomous AI Coding Agents
RigorBench evaluates AI coding agents on process discipline via five pillars and reports 41% higher process scores and 17% better outcome correctness with structured approaches on 30 tasks.
-
CoEval: Ranking Language Models for Custom Tasks Without Labeled Data or Trustworthy Benchmarks
CoEval generates task-specific benchmarks by rotating models through teacher, student, and judge roles, then weights questions by discriminative power and judges by panel consensus to recover accurate model rankings w...
-
Dataset Watermarking for Closed LLMs with Provable Detection
A new watermarking method for closed LLMs boosts random word-pair co-occurrences via rephrasing and detects the signal statistically in outputs, working reliably even when the watermarked data is only 1% of fine-tunin...
-
RoMathExam: A Longitudinal Dataset of Romanian Math Exams (1895-2025) with a Seven-Decade Core (1957-2025)
RoMathExam supplies a century-long collection of Romanian math exams together with a new intrinsic complexity metric that correlates across frontier models at r > 0.72.
-
GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models
LLMs display high variance and major accuracy drops on GSM-Symbolic variants of grade-school math problems, indicating they replicate training patterns rather than execute logical reasoning.
-
A Reference-Free Score for Detecting Silent Reasoning Failures in Large Language Models
RAFS is a reference-free composite score that flags 'silent reasoning failures'—correct answers produced by invalid chains—by combining step validity, counterfactual answer dependence, consensus, and reasoning stabili...
-
Testing Frontier Large Language Models' Physics Literacy in Parallel Physical Worlds
Introduces an auditable four-stage diagnostic for LLM physics reasoning in novel frameworks and applies it to three parallel worlds, yielding pass rates of 6/15, 6/15, and 0/15 on frontier models with noted qualitativ...
-
SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?
SWE-Bench Pro is a new benchmark with 1,865 long-horizon tasks from 41 repositories designed to evaluate AI agents on realistic enterprise-level software engineering problems beyond prior benchmarks.
-
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.
-
BitCal-TTS: Bit-Calibrated Test-Time Scaling for Quantized Reasoning Models
BitCal-TTS raises exact-match accuracy by 3.7 points (7B) and 2.8 points (14B) on small GSM8K shards for 4-bit Qwen2.5 models while cutting premature-stop rates and retaining token savings versus fixed-budget decoding.
-
The Economics of AI Training Data: A Research Agenda
The paper organizes AI training data into five exchangeable units, documents 24 licensing deals, and argues data should be a separate factor in production functions.
-
The Economics of AI Training Data: A Research Agenda
The paper synthesizes fragmented research to frame data economics around data's nonrivalry and context dependence, catalogs 2020-2025 AI training data deals, and proposes a hierarchy of data units while listing four f...
Discussion (0). Sign in to comment.