SpreadsheetBench: Towards Challenging Real World Spreadsheet Manipulation
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:3G6DGG77record.jsonopen to challenge →
read the original abstract
We introduce SpreadsheetBench, a challenging spreadsheet manipulation benchmark exclusively derived from real-world scenarios, designed to immerse current large language models (LLMs) in the actual workflow of spreadsheet users. Unlike existing benchmarks that rely on synthesized queries and simplified spreadsheet files, SpreadsheetBench is built from 912 real questions gathered from online Excel forums, which reflect the intricate needs of users. The associated spreadsheets from the forums contain a variety of tabular data such as multiple tables, non-standard relational tables, and abundant non-textual elements. Furthermore, we propose a more reliable evaluation metric akin to online judge platforms, where multiple spreadsheet files are created as test cases for each instruction, ensuring the evaluation of robust solutions capable of handling spreadsheets with varying values. Our comprehensive evaluation of various LLMs under both single-round and multi-round inference settings reveals a substantial gap between the state-of-the-art (SOTA) models and human performance, highlighting the benchmark's difficulty.
This paper has not been read by Pith yet.
Forward citations
Cited by 10 Pith papers
-
ProCUA-SFT Technical Report
ProCUA-SFT is a 3.1M-sample SFT dataset from 93K verified synthetic trajectories that lifts UI-TARS 7B OSWorld score from 26.3% to 45%.
-
Skills on the Fly: Test-Time Adaptive Skill Synthesis for LLM Agents
SkillTTA synthesizes temporary task-specific skills from retrieved training trajectories to boost LLM agent Pass@1 scores on SpreadsheetBench and BigCodeBench without parameter updates.
-
SoftSkill: Behavioral Compression for Contextual Adaptation
SoftSkill compresses agent skills into length-32 continuous prefixes via next-token training of soft deltas, yielding 5.2-12.5 point gains over SkillOpt on SearchQA and LiveMath while using far fewer tokens.
-
Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents
Insights Generator is a multi-agent system that generates evidence-backed natural-language insights characterizing systematic patterns across corpora of LLM agent execution traces.
-
Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents
Insights Generator is a multi-agent system that produces evidence-backed natural-language reports characterizing systematic patterns across populations of LLM agent execution traces.
-
ClawTrace: Cost-Aware Tracing for LLM Agent Skill Distillation
ClawTrace enables cost-aware LLM agent skill distillation by tracing per-step costs and generating preserve, prune, and repair patches, with ablations showing reduced regressions and prune rules transferring to cut co...
-
Marginal Advantage Accumulation for Memory-Driven Agent Self-Evolution
MAA formalizes alignability and comparability conditions and uses differential signals, EMA accumulation, and semantic identity merging to enable cross-batch operation-level evidence accumulation, outperforming batch-...
-
Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents
Insights Generator is a multi-agent system that produces evidence-backed insights from corpora of LLM agent traces and yields 30.4pp performance gains when humans apply the reports.
-
PoTable: Towards Systematic Thinking via Plan-then-Execute Stage Reasoning on Tables
PoTable uses multiple analytical stages with plan-then-execute code generation to produce accurate, commented, executable programs for table reasoning on WikiTQ and TabFact.
-
Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity
Seed2.0 model series reports gains in reasoning, visual understanding, search, and reliability on intricate long-horizon tasks via an internal evaluation system.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.