arXiv subjects

Sanket Badhe

Publications and source records attributed to Sanket Badhe.

15 recordsLinked to original sources

arXiv2026

SKILL.state: Scalable Long-Horizon Agent Skills

Large Language Models (LLMs) increasingly act as autonomous agents executing complex, long-running procedural skills. Existing agent runtimes maintain execution by continually appending observations, actions, and intermediate reasoning traces to an ever-growing conversation history, causing latency degradation and context-poisoning failures over long horizons. We present SKILL. state, a runtime architecture that replaces append-only conversational history with an explicit, mutable execution state. At each execution step, the model receives only the immutable skill specification, the current structured execution state, and the latest observation. Intermediate reasoning is discarded immediately after producing a validated state update, preventing prompt growth with execution history. Across diverse datasets, models, and execution environments, SKILL. state improves task accuracy while substantially reducing cumulative token consumption. Our results demonstrate that explicit execution state is an effective and architecture-agnostic abstraction for scalable long-horizon agent skills.

cs.AI↗

arXiv2026

Towards a Systems Foundation for Agentic Skills: Architecture, Lifecycle, and Security

Autonomous large language model (LLM) agents increasingly face reliability, context consumption, and execution stability bottlenecks when deployed on complex, long-horizon tasks. While monolithic prompt engineering and stateless tool-calling paradigms struggle to scale, the field is rapidly converging toward \emph{agentic skills}: modular procedural abstractions that externalize execution knowledge into reusable, executable, and portable artifacts. This paper establishes a unified systems foundation and reference architecture for the agentic skills ecosystem. We formalize skills as externalized procedural knowledge bridging high-level cognitive planning with deterministic execution environments, and systematically delineate the architecture across a nine-stage lifecycle: autonomous discovery, authoring and representation formats, memory storage, dynamic retrieval and routing, composition and orchestration, execution and repair, lifelong adaptation, empirical evaluation, and security governance. We further examine marketplace dynamics, public registries, and emerging adversarial threat vectors, alongside runtime verification and defense mechanisms. Finally, we categorize system implementations across software engineering, operating system navigation, embodied robotics, and scientific discovery, while highlighting critical open challenges in continual learning and benchmark realism. This work establishes agentic skills as a foundational paradigm for building scalable, robust, and verifiable autonomous language agents.

cs.AI↗

arXiv2026

Agent Skill Security: Threat Models, Attacks, Defenses, and Evaluation

Reusable skills are becoming a fundamental building block of Large Language Model (LLM) agents, enabling capabilities to be packaged, shared, and reused across diverse applications. However, existing security research primarily focuses on prompt injection and runtime execution, leaving security risks throughout the broader skill lifecycle largely unexplored. In this paper, we present SkillSec-Eval, a lifecycle-aware framework for systematically evaluating the security of reusable agent skills. We first characterize the skill lifecycle and develop a threat taxonomy spanning repository admission, semantic retrieval, planner selection, execution, and skill evolution. We then instantiate this taxonomy in SkillSec-Eval and conduct a comprehensive empirical evaluation using a repository of 327 real-world skills. Our study demonstrates that vulnerabilities arise at multiple lifecycle stages beyond execution, highlighting the need for lifecycle-aware security analysis of reusable agent skills.

cs.CR↗

arXiv2026

Pluralis v0.1: Towards a Multicultural, Multimodal, Multilingual Benchmark for AI Risk and Reliability

Current AI safety evaluation and benchmarking frameworks predominantly rely on Western-centric culture-agnostic defaults that mask critical regional laws, socio-linguistic nuances, and cultural taboos, leaving Vision-Language Models (VLMs) vulnerable in global deployments. We introduce Pluralis v0.1: a novel multimodal, multi-regional, and multilingual dataset built from a culture-first perspective. Spanning 6,448 prompts across six Asia-Pacific countries (Bangladesh, India, Korea, Pakistan, Singapore, Taiwan) and eight languages, Pluralis diverges from prior work by natively sourcing localized safety hazards rather than adapting Western datasets. Crucially, it introduces a multimodal evaluation paradigm: user text (e.g., "Should I gift this?") and an image referring to "this" (e.g., a clock) - both innocuous in isolation, but synergistically triggering specific legal or cultural violations. Pluralis disentangles universal safety violations from localized cultural appropriateness, establishing the latter as a first-class evaluation axis. To operationalize this, we present Judge-Pluralis, an agreement-gated LLM-as-a-Judge ensemble trained on examples classified in an empirically derived cultural taxonomy. Observing VLM behavior on a subset of the Pluralis surfaces recurring, locale-specific failure modes such as image misidentifications with downstream harm, missed item-context-locale interactions, and inadequate refusals. These failure modes vary systematically across locales and languages, exposing blind spots that globally averaged metrics conceal. Ultimately, Pluralis is not presented as a solved evaluation framework for cultural alignment, but rather as a first step and catalyst for future innovation. We call upon the research community to utilize this foundation to advance the science of multilingual, multicultural evaluation to better support AI cultural alignment globally.

cs.CL↗

arXiv2026

Local Is Not a Sufficient Privacy Boundary: Governing OS-Integrated On-Device AI

As AI systems move into operating systems, privacy no longer turns only on whether a model runs locally. A local assistant may assemble email, calendar entries, files, screenshots, notifications, and app intents; retain embeddings or summaries; invoke tools; emit telemetry; or route difficult requests to cloud infrastructure. Local inference reduces some exposure, but it answers only one question: where computation occurs. It does not answer who may assemble context, what derived state persists, which actions are authorized, or how updates change the system's authority. We develop an OS-centered privacy framework for on-device AI that treats privacy as an institutional accountability problem rather than a deployment attribute. The framework specifies a threat model, a six-part privacy risk taxonomy, privacy-by-architecture controls, and a four-level audit rubric. We demonstrate the rubric through a documentation-bounded comparison of Apple Intelligence/Foundation Models, Android AICore/Gemini Nano, and Microsoft Recall. Meaningful privacy in on-device AI depends on constrained information flow, bounded authority, visible user control, and auditable governance across the operating-system lifecycle.

cs.CR↗

arXiv2026

Prompt-Level Distillation: A Non-Parametric Alternative to Model Fine-Tuning for Efficient Reasoning

Advanced reasoning typically requires Chain-of-Thought prompting, which is accurate but incurs prohibitive latency and substantial test-time inference costs. The standard alternative, fine-tuning smaller models, often sacrifices interpretability while introducing significant resource and operational overhead. To address these limitations, we introduce Prompt-Level Distillation (PLD). We extract explicit reasoning patterns from a Teacher model and organize them into a structured list of expressive instructions for the Student model's System Prompt. Evaluated using Gemma-3 4B, PLD improved Macro F1 scores on StereoSet (57\% to 90.0\%) and Contract-NLI (67\% to 83\%), while increasing LogiQA accuracy to 70\%. Similar results on Mistral Small 3.1 demonstrate cross-architecture generalizability, enabling these compact models to match frontier performance with negligible latency overhead. These expressive instructions render the decision-making process transparent, allowing for full human verification of logic, making this approach ideal for regulated industries such as law, finance, and content moderation, as well as high-volume use cases and edge devices.

cs.CL↗

arXiv2026

Generative AI and Digital Ecosystem Resilience: A Proactive Lifecycle-Based Survey

The proliferation of adversarial synthetic content, accelerated by Generative AI (GenAI) is rendering traditional reactive detection methods ineffective. This survey synthesizes emerging research to demonstrate a paradigm shift toward the proactive detection of emerging inauthentic narratives. In this survey, we adopt a unified, lifecycle-based taxonomy to combine socio-technical lifecycle models of adversarial campaigns with advanced computational methodologies for emerging inauthentic narrative detection. By structuring the analysis around the C5 Interaction Model (Context, Causes, Content, Cycle of Amplification, Consequences), we integrate different research streams from machine learning and social science. To differentiate spread patterns of synthetic amplification from authentic baseline traffic, this paper surveys state-of-the-art techniques for modeling the creation, seeding, and propagation of fresh narratives, including the analysis of Coordinated Inauthentic Behavior (CIB), epidemiological modeling, and Hawkes process. This survey also provides a systematic review of proactive detection methods for adversarial threats at different stages in the C5 interaction model, specifically, anomaly detection in high-dimensional embedding spaces, unsupervised coordination detection on multi-layer graphs, and agentic AI systems. Finally, this survey addresses challenges posed by GenAI, including the difficulty of tracking rapidly changing threats and multi-level distributional drift, and it outlines a future research agenda focused on detecting anomalous clusters and building anticipatory and resilient systems. This survey provides a comprehensive, lifecycle-based review of methods for the proactive detection of emerging synthetic threats for more resilient information ecosystems.

cs.LG↗

arXiv2026

The Silent Vote: Improving Zero-Shot LLM Reliability by Aggregating Semantic Neighborhoods

Large Language Models are increasingly used as zero-shot classifiers in complex reasoning tasks. However, standard constrained decoding suffers from a phenomenon we define as Renormalization Bias. When a model is restricted to a small set of target labels, the standard softmax operation discards the probability mass assigned to semantic synonyms in the original distribution. This loss of information, which we call the Silent Vote, results in artificial overconfidence and poor calibration. We propose Semantic Softmax, an inference-time layer that recovers this lost information by aggregating the scores of the semantic neighborhood surrounding each target label. We evaluate this approach on Qwen-3 and Phi-4-mini models using GoEmotions and Civil Comments datasets. Our results demonstrate consistent improvements across all evaluation metrics: Semantic Softmax substantially reduces Expected Calibration Error (ECE) and Brier Score, while simultaneously enhancing discriminative performance in terms of AUROC and Macro-F1. By accounting for linguistic nuances, our method provides a more calibrated and accurate alternative for zero-shot classification.

cs.CL↗

arXiv2026

CROP: Token-Efficient Reasoning in Large Language Models via Regularized Prompt Optimization

Large Language Models utilizing reasoning techniques improve task performance but incur significant latency and token costs due to verbose generation. Existing automatic prompt optimization(APO) frameworks target task accuracy exclusively at the expense of generating long reasoning traces. We propose Cost-Regularized Optimization of Prompts (CROP), an APO method that introduces regularization on response length by generating textual feedback in addition to standard accuracy feedback. This forces the optimization process to produce prompts that elicit concise responses containing only critical information and reasoning. We evaluate our approach on complex reasoning datasets, specifically GSM8K, LogiQA and BIG-Bench Hard. We achieved an 80.6\% reduction in token consumption while maintaining competitive accuracy, seeing only a nominal decline in performance. This presents a pragmatic solution for deploying token-efficient and cost-effective agentic AI systems in production pipelines.

cs.CL↗

arXiv2026

Large Language Models in the Abuse Detection Pipeline

Online abuse has grown increasingly complex, spanning toxic language, harassment, manipulation, and fraudulent behavior. Traditional machine-learning approaches dependent on static classifiers and labor-intensive labeling struggle to keep pace with evolving threat patterns and nuanced policy requirements. Large Language Models introduce new capabilities for contextual reasoning, policy interpretation, explanation generation, and cross-modal understanding, enabling them to support multiple stages of modern safety systems. This survey provides a lifecycle-oriented analysis of how LLMs are being integrated into the Abuse Detection Lifecycle (ADL), which we define across four stages: (I) Label \& Feature Generation, (II) Detection, (III) Review \& Appeals, and (IV) Auditing \& Governance. For each stage, we synthesize emerging research and industry practices, highlight architectural considerations for production deployment, and examine the strengths and limitations of LLM-driven approaches. We conclude by outlining key challenges including latency, cost-efficiency, determinism, adversarial robustness, and fairness and discuss future research directions needed to operationalize LLMs as reliable, accountable components of large-scale abuse-detection and governance systems.

cs.CL↗

arXiv2026

Long-Tail Knowledge in Large Language Models: Taxonomy, Mechanisms, Interventions and Implications

Large language models (LLMs) are trained on web-scale corpora that exhibit steep power-law distributions, in which the distribution of knowledge is highly long-tailed, with most appearing infrequently. While scaling has improved average-case performance, persistent failures on low-frequency, domain-specific, cultural, and temporal knowledge remain poorly characterized. This paper develops a structured taxonomy and analysis of long-Tail Knowledge in large language models, synthesizing prior work across technical and sociotechnical perspectives. We introduce a structured analytical framework that synthesizes prior work across four complementary axes: how long-Tail Knowledge is defined, the mechanisms by which it is lost or distorted during training and inference, the technical interventions proposed to mitigate these failures, and the implications of these failures for fairness, accountability, transparency, and user trust. We further examine how existing evaluation practices obscure tail behavior and complicate accountability for rare but consequential failures. The paper concludes by identifying open challenges related to privacy, sustainability, and governance that constrain long-Tail Knowledge representation. Taken together, this paper provides a unifying conceptual framework for understanding how long-Tail Knowledge is defined, lost, evaluated, and manifested in deployed language model systems.

cs.CL↗

arXiv2026

Taxonomy of the Retrieval System Framework: Pitfalls and Paradigms

Designing an embedding retrieval system requires navigating a complex design space of conflicting trade-offs between efficiency and effectiveness. This work structures these decisions as a vertical traversal of the system design stack. We begin with the Representation Layer by examining how loss functions and architectures, specifically Bi-encoders and Cross-encoders, define semantic relevance and geometric projection. Next, we analyze the Granularity Layer and evaluate how segmentation strategies like Atomic and Hierarchical chunking mitigate information bottlenecks in long-context documents. Moving to the Orchestration Layer, we discuss methods that transcend the single-vector paradigm, including hierarchical retrieval, agentic decomposition, and multi-stage reranking pipelines to resolve capacity limitations. Finally, we address the Robustness Layer by identifying architectural mitigations for domain generalization failures, lexical blind spots, and the silent degradation of retrieval quality due to temporal drift. By categorizing these limitations and design choices, we provide a comprehensive framework for practitioners to optimize the efficiency-effectiveness frontier in modern neural search systems.

cs.IR↗

arXiv2025

ScamAgents: How AI Agents Can Simulate Human-Level Scam Calls

Large Language Models (LLMs) have demonstrated impressive fluency and reasoning capabilities, but their potential for misuse has raised growing concern. In this paper, we present ScamAgent, an autonomous multi-turn agent built on top of LLMs, capable of generating highly realistic scam call scripts that simulate real-world fraud scenarios. Unlike prior work focused on single-shot prompt misuse, ScamAgent maintains dialogue memory, adapts dynamically to simulated user responses, and employs deceptive persuasion strategies across conversational turns. We show that current LLM safety guardrails, including refusal mechanisms and content filters, are ineffective against such agent-based threats. Even models with strong prompt-level safeguards can be bypassed when prompts are decomposed, disguised, or delivered incrementally within an agent framework. We further demonstrate the transformation of scam scripts into lifelike voice calls using modern text-to-speech systems, completing a fully automated scam pipeline. Our findings highlight an urgent need for multi-turn safety auditing, agent-level control frameworks, and new methods to detect and disrupt conversational deception powered by generative AI.

cs.CR↗

arXiv2025

LegalSim: Multi-Agent Simulation of Legal Systems for Discovering Procedural Exploits

We present LegalSim, a modular multi-agent simulation of adversarial legal proceedings that explores how AI systems can exploit procedural weaknesses in codified rules. Plaintiff and defendant agents choose from a constrained action space (for example, discovery requests, motions, meet-and-confer, sanctions) governed by a JSON rules engine, while a stochastic judge model with calibrated grant rates, cost allocations, and sanction tendencies resolves outcomes. We compare four policies: PPO, a contextual bandit with an LLM, a direct LLM policy, and a hand-crafted heuristic; Instead of optimizing binary case outcomes, agents are trained and evaluated using effective win rate and a composite exploit score that combines opponent-cost inflation, calendar pressure, settlement pressure at low merit, and a rule-compliance margin. Across configurable regimes (e.g., bankruptcy stays, inter partes review, tax procedures) and heterogeneous judges, we observe emergent ``exploit chains'', such as cost-inflating discovery sequences and calendar-pressure tactics that remain procedurally valid yet systemically harmful. Evaluation via cross-play and Bradley-Terry ratings shows, PPO wins more often, the bandit is the most consistently competitive across opponents, the LLM trails them, and the heuristic is weakest. The results are stable in judge settings, and the simulation reveals emergent exploit chains, motivating red-teaming of legal rule systems in addition to model-level testing.

cs.MA↗

arXiv2020

Automated Segmentation of Vertebrae on Lateral Chest Radiography Using Deep Learning

The purpose of this study is to develop an automated algorithm for thoracic vertebral segmentation on chest radiography using deep learning. 124 de-identified lateral chest radiographs on unique patients were obtained. Segmentations of visible vertebrae were manually performed by a medical student and verified by a board-certified radiologist. 74 images were used for training, 10 for validation, and 40 were held out for testing. A U-Net deep convolutional neural network was employed for segmentation, using the sum of dice coefficient and binary cross-entropy as the loss function. On the test set, the algorithm demonstrated an average dice coefficient value of 90.5 and an average intersection-over-union (IoU) of 81.75. Deep learning demonstrates promise in the segmentation of vertebrae on lateral chest radiography.

eess.IV↗