Skip to content
← Back to explorer

HFEPX Hub

CS.AI + Tool Use Papers

Updated from current HFEPX corpus (Feb 27, 2026). 12 papers are grouped in this hub page. Common evaluation modes: Automatic Metrics, Simulation Env. Most common rater population: Domain Experts. Common annotation unit: Ranking. Frequently cited benchmark: Retrieval. Common metric signal: latency. Use this page to compare protocol setup, judge behavior, and labeling design decisions before running new eval experiments. Newest paper in this set is from Feb 26, 2026.

Papers: 12 Last published: Feb 26, 2026 Global RSS Tag RSS
Cs.AITool Use

Research Narrative

Grounded narrative Model: deterministic-grounded Source: persisted

Updated from current HFEPX corpus (Feb 27, 2026). This page tracks 12 papers for CS.AI + Tool Use Papers. Dominant protocol signals include automatic metrics, simulation environments, human evaluation, with frequent benchmark focus on Retrieval, BrowseComp and metric focus on latency, success rate. Use the grounded sections below to prioritize reproducible protocol choices, benchmark-matched comparisons, and judge-vs-human evaluation checks.

Why This Matters For Eval Research

Protocol Takeaways

Benchmark Interpretation

  • Retrieval appears in 16.7% of hub papers (2/12); use this cohort for benchmark-matched comparisons.
  • BrowseComp appears in 8.3% of hub papers (1/12); use this cohort for benchmark-matched comparisons.

Metric Interpretation

  • latency is reported in 16.7% of hub papers (2/12); compare with a secondary metric before ranking methods.
  • success rate is reported in 16.7% of hub papers (2/12); compare with a secondary metric before ranking methods.

Researcher Checklist

  • Close gap on Papers with explicit human feedback. Coverage is a replication risk (25% vs 45% target).
  • Close gap on Papers reporting quality controls. Coverage is a replication risk (0% vs 30% target).
  • Maintain strength on Papers naming benchmarks/datasets. Coverage is strong (50% vs 35% target).
  • Maintain strength on Papers naming evaluation metrics. Coverage is strong (50% vs 35% target).
  • Close gap on Papers with known rater population. Coverage is a replication risk (16.7% vs 35% target).
  • Close gap on Papers with known annotation unit. Coverage is a replication risk (16.7% vs 35% target).

Papers with explicit human feedback

Coverage is a replication risk (25% vs 45% target).

Papers reporting quality controls

Coverage is a replication risk (0% vs 30% target).

Papers naming benchmarks/datasets

Coverage is strong (50% vs 35% target).

Papers naming evaluation metrics

Coverage is strong (50% vs 35% target).

Papers with known rater population

Coverage is a replication risk (16.7% vs 35% target).

Papers with known annotation unit

Coverage is a replication risk (16.7% vs 35% target).

Suggested Reading Order

  1. 1. OmniGAIA: Towards Native Omni-Modal AI Agents

    Start here for detailed protocol reporting, including rater and quality-control evidence.

  2. 2. A Benchmark for Deep Information Synthesis

    Start here for detailed protocol reporting, including rater and quality-control evidence.

  3. 3. SoK: Agentic Skills -- Beyond Tool Use in LLM Agents

    Start here for detailed protocol reporting, including rater and quality-control evidence.

  4. 4. PyVision-RL: Forging Open Agentic Vision Models via RL

    Adds automatic metrics for broader coverage within this hub.

  5. 5. OpaqueToolsBench: Learning Nuances of Tool Behavior Through Interaction

    Adds simulation environments for broader coverage within this hub.

  6. 6. Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception

    Adds automatic metrics for broader coverage within this hub.

  7. 7. Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

    Adds simulation environments with pairwise preferences for broader coverage within this hub.

  8. 8. STAR: Similarity-guided Teacher-Assisted Refinement for Super-Tiny Function Calling Models

    Adds automatic metrics for broader coverage within this hub.

Known Limitations

  • Only 0% of papers report quality controls; prioritize calibration/adjudication evidence.
  • Rater population is under-specified (16.7% coverage).
  • Narrative synthesis is grounded in metadata and abstracts only; full-paper implementation details are not parsed.

Research Utility Links

human_eval vs automatic_metrics

both=1, left_only=0, right_only=8

1 papers use both Human Eval and Automatic Metrics.

automatic_metrics vs simulation_env

both=0, left_only=9, right_only=3

0 papers use both Automatic Metrics and Simulation Env.

simulation_env vs human_eval

both=0, left_only=3, right_only=1

0 papers use both Simulation Env and Human Eval.

Benchmark Brief

Retrieval

Coverage: 2 papers (16.7%)

2 papers (16.7%) mention Retrieval.

Examples: A Benchmark for Deep Information Synthesis , LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools?

Benchmark Brief

BrowseComp

Coverage: 1 papers (8.3%)

1 papers (8.3%) mention BrowseComp.

Examples: Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

Benchmark Brief

imo-answerbench

Coverage: 1 papers (8.3%)

1 papers (8.3%) mention imo-answerbench.

Examples: Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

Metric Brief

latency

Coverage: 2 papers (16.7%)

2 papers (16.7%) mention latency.

Examples: Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception , Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

Metric Brief

success rate

Coverage: 2 papers (16.7%)

2 papers (16.7%) mention success rate.

Examples: What Matters For Safety Alignment? , Measuring AI Ability to Complete Long Software Tasks

Metric Brief

cost

Coverage: 1 papers (8.3%)

1 papers (8.3%) mention cost.

Examples: Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

Most Cited In This Hub

Fast path to methods with the strongest citation traction in this scope.

Papers: OmniGAIA: Towards Native Omni-Modal AI Agents , A Benchmark for Deep Information Synthesis , SoK: Agentic Skills -- Beyond Tool Use in LLM Agents

Most Recent

Fast path to latest protocol changes and newly published evaluation setups.

Papers: OmniGAIA: Towards Native Omni-Modal AI Agents , A Benchmark for Deep Information Synthesis , SoK: Agentic Skills -- Beyond Tool Use in LLM Agents

Best Protocol Detail

Papers with explicit rater/unit metadata and quality-control signals for reproducibility.

Papers: OmniGAIA: Towards Native Omni-Modal AI Agents , A Benchmark for Deep Information Synthesis , SoK: Agentic Skills -- Beyond Tool Use in LLM Agents

Top Papers

  • OmniGAIA: Towards Native Omni-Modal AI Agents

    Xiaoxi Li, Wenxiang Jiao, Jiarui Jin, Shijian Wang, Guanting Dong · Feb 26, 2026 · Citations: 0

    Automatic Metrics Tool Use

    Human intelligence naturally intertwines omni-modal perception -- spanning vision, audio, and language -- with complex reasoning and tool usage to interact with the world.

  • A Benchmark for Deep Information Synthesis

    Debjit Paul, Daniel Murphy, Milan Gritta, Ronald Cardenas, Victor Prokhorov · Feb 24, 2026 · Citations: 0

    Human EvalAutomatic Metrics Tool Use

    Large language model (LLM)-based agents are increasingly used to solve complex tasks involving tool use, such as web browsing, code execution, and data analysis.

  • SoK: Agentic Skills -- Beyond Tool Use in LLM Agents

    Yanna Jiang, Delong Li, Haiyu Deng, Baihe Ma, Xu Wang · Feb 24, 2026 · Citations: 0

    Simulation Env Tool Use

    Agentic systems increasingly rely on reusable procedural capabilities, \textit{a.k.a., agentic skills}, to execute long-horizon workflows reliably.

  • PyVision-RL: Forging Open Agentic Vision Models via RL

    Shitian Zhao, Shaoheng Lin, Ming Li, Haoquan Zhang, Wenshuo Peng · Feb 24, 2026 · Citations: 0

    Automatic Metrics Tool Use

    Reinforcement learning for agentic multimodal models often suffers from interaction collapse, where models learn to reduce tool usage and multi-turn reasoning, limiting the benefits of agentic behavior.

  • OpaqueToolsBench: Learning Nuances of Tool Behavior Through Interaction

    Skyler Hallinan, Thejas Venkatesh, Xiang Ren, Sai Praneeth Karimireddy, Ashwin Paranjape · Feb 16, 2026 · Citations: 0

    Simulation Env Tool Use

    Tool-calling is essential for Large Language Model (LLM) agents to complete real-world tasks.

  • Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception

    Lai Wei, Liangbo He, Jun Lan, Lingzhong Dong, Yutong Cai · Feb 12, 2026 · Citations: 0

    Automatic Metrics Tool Use

    To address this, we propose Region-to-Image Distillation, which transforms zooming from an inference-time tool into a training-time primitive, thereby internalizing the benefits of agentic zooming into a single forward pass of an MLLM.

  • Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

    Ailin Huang, Ang Li, Aobo Kong, Bin Wang, Binxing Jiao · Feb 11, 2026 · Citations: 0

    Pairwise Preference Simulation Env Tool Use

    We introduce Step 3.5 Flash, a sparse Mixture-of-Experts (MoE) model that bridges frontier-level agentic intelligence and computational efficiency.

  • STAR: Similarity-guided Teacher-Assisted Refinement for Super-Tiny Function Calling Models

    Jiliang Ni, Jiachen Pu, Zhongyi Yang, Jingfeng Luo, Conggang Hu · Feb 3, 2026 · Citations: 0

    Automatic Metrics Tool Use

    The proliferation of Large Language Models (LLMs) in function calling is pivotal for creating advanced AI agents, yet their large scale hinders widespread adoption, necessitating transferring their capabilities into smaller ones.

  • What Matters For Safety Alignment?

    Xing Li, Hui-Ling Zhen, Lihao Yin, Xianzhi Yu, Zhenhua Dong · Jan 7, 2026 · Citations: 0

    Red Team Automatic Metrics Tool Use

    This paper presents a comprehensive empirical study on the safety alignment capabilities.

  • LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools?

    Guozhao Mo, Wenliang Zhong, Jiawei Chen, Qianhao Yuan, Xuanang Chen · Aug 3, 2025 · Citations: 0

    Automatic Metrics Tool Use

    Unfortunately, there is still a large gap between real-world MCP usage and current evaluation: they typically assume single-server settings and directly inject tools into the model's context, bypassing the challenges of large-scale retrieva

  • Measuring AI Ability to Complete Long Software Tasks

    Thomas Kwa, Ben West, Joel Becker, Amy Deng, Katharyn Garcia · Mar 18, 2025 · Citations: 0

    Expert Verification Automatic Metrics Tool Use

    Despite rapid progress on AI benchmarks, the real-world meaning of benchmark performance remains unclear.

  • Should You Use Your Large Language Model to Explore or Exploit?

    Keegan Harris, Aleksandrs Slivkins · Jan 31, 2025 · Citations: 0

    Automatic Metrics Tool Use

    We evaluate the ability of the current generation of large language models (LLMs) to help a decision-making agent facing an exploration-exploitation tradeoff.

Related Hubs