WorkSurface-Bench: Benchmarking Enterprise Agents on Multi-Surface Knowledge Routing

• URL: http://arxiv.org/abs/2607.25765v1
• Date: Tue, 28 Jul 2026 14:19:59 GMT
• Title: WorkSurface-Bench: Benchmarking Enterprise Agents on Multi-Surface Knowledge Routing
• Abstract summary: We introduce WorkSurface-Bench, a benchmark for evaluating this capability as surface routing.<n>It contains 1,151 atomic tasks derived from persona-scoped Workspace-Bench-Lite workspaces.<n>We evaluate four model backbones across six controlled agent settings, yielding 27,624 protocol-error-free trajectories.
• Score: 10.956110208321556
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Enterprise agents often need to integrate heterogeneous knowledge sources: documents for narrative facts, tables for computation, and dependency graphs for file relationships. Existing benchmarks typically evaluate retrieval or tool use without distinguishing whether an agent first selects the appropriate knowledge sources. We introduce WorkSurface-Bench, a benchmark for evaluating this capability as surface routing. It contains 1,151 atomic tasks derived from persona-scoped Workspace-Bench-Lite workspaces, spanning document, table, graph, and cross-surface questions. Its reference answers are auditable: table answers are reproduced through executed DuckDB queries, document answers are grounded in verified text spans, and graph answers are traced to source dependency annotations. We evaluate four model backbones across six controlled agent settings, yielding 27,624 protocol-error-free trajectories. Under gold-constrained tool access, agents achieve 98.7-99.8 Route F1, while Answer remains only 56.1-75.3 percent, showing that correct surface selection is necessary but insufficient for task completion. Matched interventions further show that surface hints improve Answer for three of four models, whereas removing irrelevant tools primarily improves routing and efficiency. In an independent three-annotator audit, all 200 sampled tasks pass all six quality criteria by majority vote, with 192 receiving unanimous judgments on every criterion. We release the dataset, construction pipeline, scoring code, and agent harness at https://github.com/haolpku/WorkSurface-Bench.

Related papers

• Evaluating Enterprise Analytics Agents: An End-to-End, Trace-Backed Methodology [10.411056951115214]
  We present an end-to-end evaluation methodology for analytics agents.<n> Grading uses question banks with human-written golden answers, repeated runs, and runtime traces.<n>We instantiate the methodology on a controlled internal analytics agent at a large online marketplace.
  arXiv  Detail & Related papers  (2026-08-28T20:57:45Z)
• VisDocAgentBench: Benchmarking Agents for Visually Rich Document Retrieval [65.89247522243959]
  VisDocAgentBench is a closed-corpus benchmark comparing static and agentic retrieval under a ranked-output contract.<n>It contains 2,375 pages from 100 documents and 120 unique-target queries balanced across direct, one-bridge, and two-bridge evidence structures.<n>A strong late-interaction visual retriever reaches 97.50% Recall@1 on direct items but 2.50% on two-bridge items, exposing the limits of query--target matching when relevance depends on corpus context.
  arXiv  Detail & Related papers  (2026-08-18T15:23:06Z)
• Schema-Agnostic Graph Reasoning Agent for Hybrid Knowledge Graphs [42.698418800007865]
  We present GRA, a Graph Reasoning Agent that explores hybrid knowledge graphs.<n>On UFK-M, an industrial benchmark of 258 analytical questions, GRA beats a full-context agent by 5.1 pp.<n>A graph-free control shows the gain comes chiefly from selective agentic access rather than graph topology.
  arXiv  Detail & Related papers  (2026-08-16T16:10:32Z)
• Learning Compositional Meta-Routing for Agentic Workflows: An Executable Benchmark [0.0]
  We introduce an executable benchmark and a budget-aware meta-router that composes heterogeneous operations from raw task text.<n>Results: learned success falls to 75.9% and trails static routing at 93.5%, while remaining 49% cheaper and exceeding one-shot routing by 34.3 points.
  arXiv  Detail & Related papers  (2026-07-31T07:00:49Z)
• Tencent WorkBuddy Bench: A Multi-Domain Coding-Agent Benchmark with Contamination-Resistant Task Construction [57.138342889101345]
  Tencent WorkBuddy Bench is a multi-domain evaluation suite for coding agents.<n>This report documents its construction methodology, scoring protocol, and a cross-model leaderboard.
  arXiv  Detail & Related papers  (2026-07-23T04:34:06Z)
• SGR-Bench: Benchmarking Search Agents on State-Gated Retrieval [5.32888938421653]
  A class of specialized retrieval tasks remains undercharacterized.<n>We term this capability state-gated retrieval (SGR)<n>SGR-Bench is a benchmark for this setting containing 100 expert-curated tasks spanning six source families and 12 public data ecosystems.
  arXiv  Detail & Related papers  (2026-05-21T09:22:48Z)
• Towards Multi-Agent Autonomous Reasoning in Hydrodynamics [0.06999740786886537]
  We present a multi-agent system (MAS) prototype for hydrodynamics in which specialized agents are coordinated through a Layer Execution Graph (LEG)<n>A planner agent constructs query-specific execution topologies from natural-language routings that capture domain knowledge without hard-coding it as rigid control logic.<n>A reporter agent synthesizes the final response, while the runtime logs provenance for every tool invocation to support auditability.
  arXiv  Detail & Related papers  (2026-05-01T21:17:55Z)
• TADI: Tool-Augmented Drilling Intelligence via Agentic LLM Orchestration over Heterogeneous Wellsite Data [0.0]
  TADI (Tool-Augmented Drilling Intelligence) is an agentic AI system that transforms drilling operational data into evidence-based analytical intelligence.<n>TADI integrates 1,759 daily drilling reports, selected WIT real-time objects, 15,634 production records, formation tops, and perforations into a dual-store architecture.
  arXiv  Detail & Related papers  (2026-04-30T03:19:39Z)
• ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents [77.22389710754452]
  We introduce bench, a benchmark for coworker agents built around multi-turn multi-day tasks.<n>The current release contains 100 tasks across 13 professional scenarios, executed against five stateful sandboxed services.<n>The strongest model reaches 75.8 weighted score, but the best strict Task Success is only 20.0%, indicating that partial progress is common.
  arXiv  Detail & Related papers  (2026-04-26T16:05:02Z)
• Visual Document Understanding and Question Answering: A Multi-Agent Collaboration Framework with Test-Time Scaling [83.78874399606379]
  We propose MACT, a Multi-Agent Collaboration framework with Test-Time scaling.<n>It comprises four distinct small-scale agents, with clearly defined roles and effective collaboration.<n>It shows superior performance with a smaller parameter scale without sacrificing the ability of general and mathematical tasks.
  arXiv  Detail & Related papers  (2025-08-05T12:52:09Z)
• RAG over Tables: Hierarchical Memory Index, Multi-Stage Retrieval, and Benchmarking [63.253294691180635]
  In real-world scenarios, beyond pure text, a substantial amount of knowledge is stored in tables.<n>We first propose a table-corpora-aware RAG framework, named T-RAG, which consists of the hierarchical memory index, multi-stage retrieval, and graph-aware prompting.
  arXiv  Detail & Related papers  (2025-04-02T04:24:41Z)
• KET-QA: A Dataset for Knowledge Enhanced Table Question Answering [63.56707527868466]
  We propose to use a knowledge base (KB) as the external knowledge source for TableQA. Every question requires the integration of information from both the table and the sub-graph to be answered. We design a retriever-reasoner structured pipeline model to extract pertinent information from the vast knowledge sub-graph.
  arXiv  Detail & Related papers  (2024-05-13T18:26:32Z)
• TANQ: An open domain dataset of table answered questions [15.323690523538572]
  TANQ is the first open domain question answering dataset where the answers require building tables from information across multiple sources.<n>We release the full source attribution for every cell in the resulting table and benchmark state-of-the-art language models in open, oracle, and closed book setups.<n>Our best-performing baseline, Gemini Flash reaches an overall F1 score of 60.7, lagging behind human performance by 12.3 points.
  arXiv  Detail & Related papers  (2024-05-13T14:07:20Z)
• KILT: a Benchmark for Knowledge Intensive Language Tasks [102.33046195554886]
  We present a benchmark for knowledge-intensive language tasks (KILT) All tasks in KILT are grounded in the same snapshot of Wikipedia. We find that a shared dense vector index coupled with a seq2seq model is a strong baseline.
  arXiv  Detail & Related papers  (2020-09-04T15:32:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.