arXiv subjects

Melveena Jolly

Publications and source records attributed to Melveena Jolly.

4 recordsLinked to original sources

arXiv2026

Stabilized Best-of-$K$ Training for Neural Combinatorial Optimization

Leader Reward modifies POMO training to emphasize the best trajectory produced by repeated inference. We test a narrow extension: replace its binary leader/non-leader distinction with a stabilized rank signal indexed by a sampling budget $K$. With the POMO architecture, 3,050-epoch schedule, and TSP-100 test set held fixed, the Leader Reward reimplementation obtains $7.7662$ under 100-start, 8-augmentation greedy decoding, matching the reported $7.766$ at its displayed precision. Under independent sampling, the stabilized $K=8$ recipe lowers realized Best-of-8 cost in all three paired training seeds: $7.7944$ versus $7.8136$. This observation is estimation-only and decoder-specific: three seeds are below the six-seed testing floor, Leader Reward is better at sampled $K=1$, and it remains slightly better under its original augmented-greedy protocol. We make no unbiased-estimator, universal superiority, or state-of-the-art claim.

cs.LG↗

arXiv2026

Rank-Conditioned Sample Reuse for the Plackett--Luce Best-of-$K$ Objective

We study the coupled objective J_K^WOR = E_{S ~ PL-WOR_K}[max_{i in S} R_i]: the expected maximum reward of a size-K Plackett-Luce draw without replacement, the law of Gumbel-Top-K / Stochastic Beam Search decoding. This estimand differs from the conventional i.i.d. objective J_K^iid = E[max_{i<=K} R_i] targeted by existing sample-reuse Max@K estimators, and reusing their i.i.d. weights under the coupled sampler is provably biased (a closed-form three-item instance gives E[g_iid] = (4/5) grad J_K^WOR exactly; pass@K under the coupled sampler is the binary-reward special case). Generic joint-score REINFORCE is already unbiased for J_K^WOR; what it lacks is sample reuse. Our contribution is to instantiate standard rank-conditioned Horvitz-Thompson estimation for the J_K^WOR subset total: from one Gumbel-Top-n pool (n>K) and its observed priority threshold we build an estimator that reuses all C(n,K) embedded K-subsets, unbiased with an unbiased exact score-function surrogate gradient, plus a reward-sorted Max-specific dynamic program that collapses the C(n,K)-term subset sum (with K!-cost set probabilities) exactly to a one-dimensional integral. A fixed-Q quadrature evaluation costs O(n log n + nKQ) arithmetic and is numerically, not algebraically, exact; no epsilon-approximation rate is certified. Each nonzero degree-K Horvitz-Thompson term has finite second moment exactly when n >= 2K; under the same assumptions the full surrogate gradient has finite second moment whenever n >= 2K (sharpness there is open). At K=1 the construction recovers classical priority sampling. All quantities require only the values and differentiable computation graphs of the n+1 drawn items' probabilities, so finite structured sequence policies sampled by exact SBS are covered. A certified finite-Q quadrature bound and countably infinite support remain open. Validation code is included as ancillary files.

cs.LG↗

arXiv2026

IndustriConnect: MCP Adapters and Mock-First Evaluation for AI-Assisted Industrial Operations

AI assistants can decompose multi-step workflows, but they do not natively speak industrial protocols such as Modbus, MQTT/Sparkplug B, or OPC UA, so this paper presents INDUSTRICONNECT, a prototype suite of Model Context Protocol (MCP) adapters that expose industrial operations as schema-discoverable AI tools while preserving protocol-specific connectivity and safety controls; the system uses a common response envelope and a mock-first workflow so adapter behavior can be exercised locally before connecting to plant equipment, and a deterministic benchmark covering normal, fault-injected, stress, and recovery scenarios evaluates the flagship adapters, comprising 870 runs (480 normal, 210 fault-injected, 120 stress, 60 recovery trials) and 2820 tool calls across 7 fault scenarios and 12 stress scenarios, where the normal suite achieved full success, the fault suite confirmed structured error handling with adapter-level uint16 range validation, the stress suite identified concurrency boundaries, and same-session recovery after endpoint restart is demonstrated for all three protocols, with results providing evidence spanning adapter correctness, concurrency behavior, and structured error handling for AI-assisted industrial operations.

cs.SE↗

arXiv2026

Agentproof: Static Verification of Agent Workflow Graphs

Agent frameworks increasingly encode tool-using behavior as explicit workflow graphs, yet safety enforcement remains a runtime concern. These frameworks expose analyzable graph structure through their APIs, enabling pre-deployment static verification of safety properties that runtime guardrails can only check reactively. This paper presents Agentproof, a system that automatically extracts a unified abstract graph model from four major agent frameworks (LangGraph, CrewAI, AutoGen, Google ADK), applies six structural checks with witness trace generation, and evaluates temporal safety policies via a DSL compiled to deterministic finite automata, both statically through a graph x DFA product construction and at runtime over event traces. Unlike general-purpose model checkers, Agentproof requires no manual modeling. In a curated benchmark of 18 author-constructed workflows, 27% of the benchmark contain structural defects (dead-end nodes, unreachable exits) and 55% violate a human-gate policy when enforced, distinct categories that prior work conflates. All 15 temporal policies defined fit within the seven-form DSL fragment, and verification completes in sub-second time for graphs up to 5,000 nodes. The corpus serves as a reproducible benchmark for evaluating static verification tools rather than as a prevalence study; defect rates reflect tool detection capability on a targeted benchmark, not base rates in production systems. Nonetheless, static graph verification complements runtime guardrails by catching topology-level defects that runtime tools miss unless the offending path is exercised.

cs.LO↗