{ "schema_version": 1, "builder_plan": { "scenario": "An assistant chooses the wrong API or sends arguments your product cannot use.", "change": "Apply the proposed tool-selection or argument-generation change in a sandbox.", "baseline": "Your current agent with the same tool schemas and task budget.", "steps": [ "Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.", "Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.", "Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.", "Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test." ], "measure": "Correct completed actions, invalid calls, retries, latency, and cost per completed task.", "decision": "Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation." }, "builder_review": null, "arxiv_id": "2610.21449", "title": "Incentivizing Tool-Use Reinforcement Learning for Post-Arrival Emergency Care Prediction", "summary": "The method replaces prompt-only LLM reasoning and standalone gradient-boosted trees for tabular-plus-text EHR prediction with an RL-trained 4B agent that calls feature groups as tools, using rewards derived from a gradient-boosted tree teacher and a reverse curriculum that distills clinical skills from the prompt into the weights. The cost is a full RL training pipeline (teacher model, reward engineering, curriculum schedule) plus clinical data access, which is feasible at 4B scale only with parameter-efficient fine-tuning on a single 24 GB GPU and remains operationally heavier than simply deploying a tree model. What can fail: reward hacking against the tree teacher, degraded performance on institutions whose feature groups differ from the training EHR schema, and failure to transfer to non-clinical tabular tasks where the teacher-curriculum pipeline must be rebuilt.", "rationale": "The approach requires RL training infrastructure, a task-specific teacher model, and domain EHR data that the constrained reader is unlikely to justify building when conventional gradient-boosted trees already provide a strong, cheap baseline for tabular clinical prediction.", "technique": "Tool-integrated RL with tree-teacher rewards for EHR prediction", "family": "uso_de_ferramenta", "recommendation": "observar", "published": "2026-10-15", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2610.21449", "independent_implementations": 2, "total_implementations": 3, "stars": 136, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "The method replaces prompt-only LLM reasoning and standalone gradient-boosted trees for tabular-plus-text EHR prediction with an RL-trained 4B agent that calls feature groups as tools, using rewards derived from a gradient-boosted tree teacher and a reverse curriculum that distills clinical skills from the prompt into the weights. The cost is a full RL training pipeline (teacher model, reward engineering, curriculum schedule) plus clinical data access, which is feasible at 4B scale only with parameter-efficient fine-tuning on a single 24 GB GPU and remains operationally heavier than simply deploying a tree model. What can fail: reward hacking against the tree teacher, degraded performance on institutions whose feature groups differ from the training EHR schema, and failure to transfer to non-clinical tabular tasks where the teacher-curriculum pipeline must be rebuilt.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.21449", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "A 4B agent achieves the highest macro F1 across compared models on three clinical prediction tasks, outperforming conventional ML baselines and remaining competitive with much larger frontier LLMs; no multiplicative factor is reported.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.21449", "source_page": null, "source_excerpt": "" } ], "exposure_map": [ { "dimension": "quality", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compute", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "latency", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "operations", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compatibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "security", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "data_and_training", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "reproducibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" } ], "risks": [], "minimum_test": [], "open_questions": [], "independent_tests": [], "lab_reproductions": [], "report_available": false, "validation_tier": "unknown", "evidence_tier": "unknown", "infrastructure_basis": "unknown", "equations": [], "equations_status": "not_fetched", "core_kind": "", "equations_fetched_at": "", "equations_source_url": "", "equations_source_sha256": "", "exposures_reviewed_at": "", "exposures_source_sha256": "" }