Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

ProCompNav: Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries

Junhyuk Kwon, Seungjoon Lee, Hyejin Park, Kyle Min +1 more

Published

May 7, 2026

Citations

0

Trust level

Moderate

Usefulness score

25/100 (Low)

Extraction confidence

55% (Moderate)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Sep 29, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics General Web Browsing

Should you rely on this paper?

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Use this for comparison and orientation, not as your only source.

Best use

Background context only

Use if you need

A benchmark-and-metrics comparison anchor.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

No major weakness surfaced.

Human feedback signal
Not explicit
Not explicit in abstract metadata
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
25/100
Adjacent candidate

Treat as adjacent context, not a core eval-method reference.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance. A practical agent should reduce the user's burden by actively asking only the information needed to distinguish the target from similar distractors, rather than requiring a detailed description upfront. Existing approaches often fall short of this goal by mistaking distractors that strongly match the accumulated information about the target provided by the user. As a result, despite the dialogue, the agent may still fail to distinguish the target from distractors, leading to premature decisions and lengthy user responses. We propose Proactive Instance Navigation with Comparative Judgment (ProCompNav), a two-stage framework that first constructs a candidate pool and then identifies the target through Recursive Comparative Judgment (RCJ). RCJ iteratively narrows the pool by selecting an attribute-value pair that divides the candidates, asking the user a binary question, and removing inconsistent candidates, without requiring an attribute unique to the target. On CoIN-Bench, ProCompNav outperforms the evaluated baselines in Success Rate while substantially reducing Response Length. On the non-interactive TextNav benchmark, ProCompNav achieves the highest Success Rate. Two human studies further show that participants prefer ProCompNav's interaction strategies.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

missing

None explicit

No explicit feedback protocol extracted.

"Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance."

Evaluation Modes

strong

Automatic Metrics

Includes extracted eval setup.

"Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance."

Benchmarks / Datasets

strong

Coin Bench

Useful for quick benchmark comparison.

"On CoIN-Bench, ProCompNav outperforms the evaluated baselines in Success Rate while substantially reducing Response Length."

Reported Metrics

strong

Success rate

Useful for evaluation criteria comparison.

"On CoIN-Bench, ProCompNav outperforms the evaluated baselines in Success Rate while substantially reducing Response Length."

Benchmarks and datasets

Coin-Bench

Reported metrics

success rate
Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Not reported
Expertise required
General
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
Web Browsing
Quality controls
Not reported
Evidence quality
Moderate
Use this page as
Background context only

Research brief

Metadata summary

Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • Natural-language instance navigation becomes challenging when the initial user request does not uniquely specify the target instance.
  • A practical agent should reduce the user's burden by actively asking only the information needed to distinguish the target from similar distractors, rather than requiring a detailed description upfront.
  • Existing approaches often fall short of this goal by mistaking distractors that strongly match the accumulated information about the target provided by the user.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Check the full text for explicit evaluation design choices (raters, protocol, and metrics).
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Contribution summary

  • A practical agent should reduce the user's burden by actively asking only the information needed to distinguish the target from similar distractors, rather than requiring a detailed description upfront.
  • As a result, despite the dialogue, the agent may still fail to distinguish the target from distractors, leading to premature decisions and lengthy user responses.
  • We propose Proactive Instance Navigation with Comparative Judgment (ProCompNav), a two-stage framework that first constructs a candidate pool and then identifies the target through Recursive Comparative Judgment (RCJ).

Why it matters for eval

  • A practical agent should reduce the user's burden by actively asking only the information needed to distinguish the target from similar distractors, rather than requiring a detailed description upfront.
  • As a result, despite the dialogue, the agent may still fail to distinguish the target from distractors, leading to premature decisions and lengthy user responses.

Researcher checklist

  • Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Evaluation mode is explicit

    Detected: Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    Detected: Coin-Bench

  • Metric reporting is present

    Detected: success rate