Pith. sign in

REVIEW 16 cited by

DSBench: How Far Are Data Science Agents from Becoming Data Science Experts?

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 2409.07703 v3 pith:5S7QIIHZ submitted 2024-09-12 cs.AI cs.CL

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

Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) have demonstrated impressive language/vision reasoning abilities, igniting the recent trend of building agents for targeted applications such as shopping assistants or AI software engineers. Recently, many data science benchmarks have been proposed to investigate their performance in the data science domain. However, existing data science benchmarks still fall short when compared to real-world data science applications due to their simplified settings. To bridge this gap, we introduce DSBench, a comprehensive benchmark designed to evaluate data science agents with realistic tasks. This benchmark includes 466 data analysis tasks and 74 data modeling tasks, sourced from Eloquence and Kaggle competitions. DSBench offers a realistic setting by encompassing long contexts, multimodal task backgrounds, reasoning with large data files and multi-table structures, and performing end-to-end data modeling tasks. Our evaluation of state-of-the-art LLMs, LVLMs, and agents shows that they struggle with most tasks, with the best agent solving only 34.12% of data analysis tasks and achieving a 34.74% Relative Performance Gap (RPG). These findings underscore the need for further advancements in developing more practical, intelligent, and autonomous data science agents.

Discussion (0). Sign in to comment.

Forward citations

Cited by 16 Pith papers

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

  1. Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories

    cs.CV 2026-06 unverdicted novelty 7.0 of 10

    Data2Story is a multi-agent framework that generates evidence-grounded multimodal articles from data, evaluated on 18 articles against human pieces for verifiability, angle coverage, and quality across human, rubric, ...

  2. FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics

    cs.LG 2026-05 unverdicted novelty 7.0 of 10

    FML-Bench shows a simple greedy hill-climber nearly matches tree search on dense-opportunity tasks while an adaptive agent that broadens search on stagnation outperforms six baselines across 18 tasks.

  3. KompeteAI: Accelerated Autonomous Multi-Agent System for End-to-End Pipeline Generation for Machine Learning Problems

    cs.AI 2025-08 unverdicted novelty 7.0 of 10

    KompeteAI accelerates AutoML pipeline evaluation 6.9 times and beats prior systems by 3% on MLE-Bench through candidate merging, external RAG, and predictive early scoring.

  4. MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering

    cs.CL 2024-10 unverdicted novelty 7.0 of 10

    MLE-bench evaluates frontier language models as ML engineering agents on 75 Kaggle competitions, with the top setup (o1-preview + AIDE) reaching bronze medal level in 16.9% of tasks.

  5. CausalGame: Benchmarking Causal Thinking of LLM Agents in Games

    cs.CL 2026-07 conditional novelty 6.5 of 10

    No evaluated frontier LLM agent shows reliable causal thinking on CausalGame’s 14 SCM-based drone games; survival and mechanistic understanding remain largely decoupled.

  6. ExplainBench: Evaluating Code Explanations from Agents

    cs.SE 2026-07 conditional novelty 6.0 of 10

    Agent explanation quality can be scored automatically with LLM questionnaires, and this score ranks agents differently from SWE-bench Verified.

  7. Before the Action: Benchmarking LLMs on Prospective Hypothesis Discovery

    cs.CL 2026-07 conditional novelty 6.0 of 10

    HypoArena is a 988-case benchmark that asks LLMs to generate hypothesis sets from conclusion-free reconstructed contexts and ranks 15 models via pairwise arena judgments.

  8. FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics

    cs.LG 2026-05 accept novelty 6.0 of 10

    FML-Bench shows that a simple greedy hill-climber performs nearly as well as complex tree-search agents on ML research tasks, with an adaptive strategy that switches exploration modes outperforming all tested agents.

  9. Pioneer Agent: Continual Improvement of Small Language Models in Production

    cs.AI 2026-04 unverdicted novelty 6.0 of 10

    Pioneer Agent automates the full lifecycle of adapting and continually improving small language models via diagnosis-driven data synthesis and regression-constrained retraining, delivering gains of 1.6-83.8 points on ...

  10. StatEval: A Comprehensive Benchmark for Large Language Models in Statistics

    cs.CL 2025-10 conditional novelty 6.0 of 10

    StatEval is a new 16,000-question statistics benchmark showing that even strong LLMs score below 60% on research-level statistical proof tasks.

  11. When Should Users Check? Modeling Confirmation Frequency inMulti-Step Agentic AI Tasks

    cs.HC 2025-10 conditional novelty 6.0 of 10

    A decision-theoretic model based on the observed Confirmation-Diagnosis-Correction-Redo user pattern places intermediate confirmations in AI agent tasks, yielding 81% user preference and 13.54% faster completion versu...

  12. RExBench: Can coding agents autonomously implement AI research extensions?

    cs.CL 2025-06 unverdicted novelty 6.0 of 10

    RExBench is a new benchmark showing that LLM coding agents fail to autonomously implement most realistic research extensions to prior AI papers.

  13. Matryoshka Agent: Unfolding Sub-Agents for Long-Horizon Machine Learning Engineering

    cs.AI 2026-07 conditional novelty 5.0 of 10

    Matryoshka Agent’s orchestrator–sub-agent hierarchy plus tree-ranked RL raises MLE-Dojo HumanRank, letting a 4B orchestrator approach o4-mini and giving a 30B coder up to 36.7% relative gain.

  14. Business Utility of Large Language Models as Exploratory Data Analysis Agents

    cs.CY 2026-05 unverdicted novelty 5.0 of 10

    Evaluation of 15 LLM configurations across four conditions in a supply chain EDA benchmark finds most lack sufficient repeatability for autonomous deployment, with GPT-5.4 at extra-high reasoning effort scoring highes...

  15. CIPHER: A Decoupled Exploration-Selection Framework for Test-Time Scaling of Data Science Agents

    cs.AI 2026-07 conditional novelty 4.0 of 10

    A decoupled exploration-selection framework for test-time scaling lifts data-science agent accuracy by up to roughly 12 percentage points over matched baselines.

  16. How Far Are AI Scientists from Changing the World?

    cs.AI 2025-07 conditional novelty 4.0 of 10

    This survey proposes a four-level capability framework for AI Scientist systems and, using an AI reviewer, finds that current systems produce papers rated well below normal scientific standards.

Pith tools