Searcharxiv⌕ Search

arXiv subjects

Bhaskar Gurram

Publications and source records attributed to Bhaskar Gurram.

3 recordsLinked to original sources

Seeing Is Not Believing: Counterfactual Lesion Erasure Teaches a Small Medical Vision-Language Agent to Report Only What the Image Shows

Medical vision-language models (VLMs) often report a finding because the question names it; standard benchmarks miss this. From public segmentation datasets we build counterfactual views without manual annotation: the lesion erased by inpainting, a same-shape control edit elsewhere, and unedited lesion-free images. The evidence sensitivity gap (ESG) is the rate of absent-or-refer answers after erasure minus the rate after the control edit; it is zero for a model that ignores the image or reacts to both edits alike. MedScope-4B (Qwen3.5-4B, LoRA, image tools) was trained on label-derived trajectories, selected by a pre-registered rule and tested once against nine frontier and six open VLMs; four pre-registered checks then probed the mechanism. On 507 internal cases across five imaging types, MedScope-4B reached ESG 0.758 and claimed an erased lesion in 9.6% of cases, against 0.27-0.47 and 41-62% for the frontier models (all p<0.001), without significant loss of accuracy on real lesions. Without counterfactual views ESG fell to 0.279+-0.107 (three seeds); the effect replicated on MedGemma 1.5 4B and two unseen inpainting tools. At four external centres the advantage held but frontier models detected more real lesions; on a fresh external test MedScope-4B detected 0 of 48 pneumothorax films, with net utility below gemini-4-argon's (-0.36 [-0.50, -0.23]). Matching the control edit was not shown to matter; over three seeds no fresh-external ESG advantage over removal-based DPO was established; a paired-view consistency objective did not beat a matched-compute control, while that continuation alone cut erased-lesion claims on unseen tools at every seed, at a cost in sensitivity at one. Training on erased and edited images gives a small model an evidence-gated disposition that prompting does not give frontier models; it does not improve perception, and rare critical findings need coverage.

cs.AI↗

Measuring What Tool-Using LLM Agents Actually Do: Structured Validation, Causal Error Propagation, and Fabricated Tool Use

Tool-using agents built on large language models (LLMs) are usually scored by comparing the final answer with a reference text, using a string heuristic or an LLM judge. This is unreliable when the references describe the task rather than its answer. We replace it with a deterministic validator that checks the tools called, the executed parameters, the resulting state, and whether the values in the answer are supported by the correct tool observations. We re-scored 33,583 traces from 13 agents on simulated calendar, weather, medical and retail tools. The simulator reproduced all 10,396 logged observations, and the gold observations supported the validator in 16 of 18 disagreements with a three-annotator majority (the annotators in none). The instrument changes conclusions: on the same 306 traces, Gemini-2.0-Flash scores 63.4% under the substring heuristic and 5.6% under the validator. In a paired design (same task and model, with and without one schema-valid wrong argument injected into the first tool call), the error rate rises from 52.4% to 74.5% (+22.0 percentage points, 95% CI 18.4-25.5; 1,085 pairs, 7 models), and by 44.7 points (38.4-51.0) where the corrupted call executed without errors. An LLM-judge ensemble estimates +2.5 points on the same pairs. Agents also report tool results without executing any call, in 0-42.5% of traces depending on the model (rule-based lower bounds), largely because the harness accepts an action and a final answer in one completion. A rule-based runtime monitor costs about 7 microseconds per trace but reduces wrong answers mainly by abstaining.

cs.AI↗

Valid Per-Field Selective Risk Control for Document Extraction: Three Failure Modes, a Validity Ladder, and When Conditioning Pays

Per-field accept/review with selective risk at most alpha -- accept a field only if the error rate among accepted fields is controlled -- is the trust contract document-extraction systems need, and the natural procedure silently violates it on real documents. On 13,859 genuine claude-sonnet-5 fields from 800 CORD receipts (49.0% correct) we diagnose three failure modes: document clustering (design effect 1.84-2.45), score-refit leakage (coverage 0.416 at risk 0.127, violating alpha=0.10 in 95% of splits), and a tie-mass pathology (a degenerate score collapses the threshold grid, 0.030 to 0.001). We organize the fixes as a validity ladder, guarantee form stated per tier. A fit/val split protocol restores expected-selective-risk control for a learned fusion: coverage 0.318 at risk 0.096 at nominal alpha=0.10, no tolerance band (production variant 0.326) -- an on-average point whose realized risk exceeds alpha in 47.5% of resplits, not a certificate. Mondrian Learn-then-Test with exact binomial tails yields per-group PAC certificates: field-iid 0.171 at risk 0.068, cluster-corrected 0.140, doc-iid 0.060 -- the only tier matching documents, honestly near-vacuous today. Support-bin, the pre-specified provenance taxonomy, wins every rigor tier on the sonnet CORD capture (p<1e-4, Bonferroni-corrected) -- a win that does not replicate on the same documents under haiku or qwen -- while on higher-accuracy corpora pooled thresholds win: conditioning helps exactly where pooled cannot certify, subsumed by a learned score elsewhere. A frozen-configuration confirmation on selection-untouched claude-haiku-4-5 held at both risk levels, and a blind three-annotator human-gold audit verifies the practical tier's accepted-set risk at 1.3% against its 10% budget (Fleiss' kappa=0.83; labels err one-sidedly pessimistic). Released Apache-2.0 with seed-pinned, regression-gated procedures.

cs.LG↗