Cost-aware stopping over ranked tool prefixes, trained on stop-versus-continue payoff gaps, beats score and sufficiency thresholds under heterogeneous tool costs.
Jaewan Park, Solbee Cho, and Jay-Yoon Lee
6 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
6
Pith papers citing it
1
external citations · external index
citation-role summary
background 1
citation-polarity summary
years
2026 6roles
background 1polarities
background 1representative citing papers
On ChemKGMultiHopQA, iterative retrieval-reasoning outperformed oracle gold-context static RAG for all 11 LLMs tested, with gains up to 25.6 percentage points.
SAGE uses offline imitation learning from a latency-quality oracle to choose per-query retrieval budgets, reporting 95% SLO compliance and 51% lower retrieval cost on Natural Questions with 2 points of EM loss.
TASR provides a training-free predicate that stops iterative retrieval on repeated normalized answers plus calibrated logit margin above 0.25, retaining 94.8% of fixed-k=5 F1 at 62.6% of the calls across 32 configurations.
Reporting a local decision metric and a task outcome in the same paper does not establish that better local decisions caused the task-level gain; only matched, controlled comparisons can.
citing papers explorer
-
Scores Are Not Decisions: Cost-Aware Stopping for Tool Acquisition in LLM Agents
Cost-aware stopping over ranked tool prefixes, trained on stop-versus-continue payoff gaps, beats score and sufficiency thresholds under heterogeneous tool costs.
-
When Iterative RAG Beats Ideal Evidence: A Diagnostic Study in Scientific Multi-hop Question Answering
On ChemKGMultiHopQA, iterative retrieval-reasoning outperformed oracle gold-context static RAG for all 11 LLMs tested, with gains up to 25.6 percentage points.
-
SAGE: SLO-Aware Adaptive Retrieval for Production RAG Systems
SAGE uses offline imitation learning from a latency-quality oracle to choose per-query retrieval budgets, reporting 95% SLO compliance and 51% lower retrieval cost on Natural Questions with 2 points of EM loss.
-
TASR: Training-Free Adaptive Stopping for Iterative Retrieval
TASR provides a training-free predicate that stops iterative retrieval on repeated normalized answers plus calibrated logit margin above 0.25, retaining 94.8% of fixed-k=5 F1 at 62.6% of the calls across 32 configurations.
-
What Can Component-Replacement Evidence Establish? A Critical Scoping Review of Local Decisions in LLM Agents
Reporting a local decision metric and a task outcome in the same paper does not establish that better local decisions caused the task-level gain; only matched, controlled comparisons can.
- HALT: Verification-Aware Stopping for Retrieval-Augmented Search Agents