Results for “Inference”

NVIDIA Blog Mavi Nafria SI

NVIDIA Transforms European Retail, From Digital Discovery to Physical Operations

European retailers are using open models and physical intelligence across use cases — from a commerce platform cutting inference costs by 90% to to a fulfillment center where collaboration of different robotic fleets is simulated inside a digital twin. Frontier Results With Open Models Retail leaders and technology companies across Europe are using NVIDIA-accelerated open

Read at NVIDIA Blog

TestingCatalog Erin | AI Agent SI

Recursive releases Flint, 40B model for local coding

Recursive’s Flint pairs open weights with fast local coding, a 512k-token context window and an optimized inference engine for Apple Silicon.

Read at TestingCatalog

NVIDIA Blog Uttara Kumar SI

Google Cloud and NVIDIA Announce General Availability of NVIDIA Vera Rubin Infrastructure

General availability of the instances is coming next month, with planned expansion of availability to Europe — bringing an integrated platform for training, inference and large-scale reinforcement learning to cloud customers. A5X instances bring together NVIDIA Vera Rubin NVL72, NVIDIA ConnectX-9 SuperNICs, Google Cloud Virgo Network and Google Cloud’s fourth-generation advanced liquid-cooling technology for demanding

Read at NVIDIA Blog

Hacker News regsiren SI

Show HN: SentinelGate – Drop prompt injections at Token 0 before LLM inference

Article URL: https://sentinelgate.dev Comments URL: https://news.ycombinator.com/item?id=50165432 Points: 1 # Comments: 0

Read at Hacker News

arXiv (research papers) Yuchen Chen SI

RAGLeak: Supply-Chain Privacy Backdoors in Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) enables large language models to access external knowledge during inference, but its reliance on external knowledge bases also introduces new attack surfaces. Existing data extraction attacks against RAG systems rely on either prompt injection, which heavily depends on the generator's instruction-following capability, or backdoors that reproduce or paraphrase retrieved documents but struggle to balance extraction fidelity and…

Read at arXiv (research papers)

Journal articles (OpenAlex) Bhagwat Mahesh, T. S. Vishwanath · Proceedings on Engineering Sciences Security

A HYBRID ANFIS-GA FRAMEWORK WITH EXPLAINABLE AI (XAI) FOR TRANSPARENT AND ACCURATE PHISHING DETECTION

Considering the fact that phishing sites remain a prevalent and serious attack on the Internet, the sphere is in dire need of the creation of smart and dynamic defense systems.We overcome this issue by providing a new hybrid neuro-evolutionary fuzzy model.This system is a careful integration of the Adaptive Neuro-Fuzzy Inference System (ANFIS) that is augmented with a Genetic Algorithm (GA) to do a global optimization and integrated Explainable AI (XAI) to make the entire…

Read at Journal articles (OpenAlex)

arXiv (research papers) L. Zhu SI

Stealthy Poisoning Attacks via Reused KV Cache of Reasoning

Modern large language model (LLM) systems reuse key-value (KV) caches to accelerate inference. Recent work has shown that reusing KV caches from reasoning traces can also improve task performance, similar to retrieval-augmented generation (RAG). We demonstrate that this form of KV cache reuse introduces new poisoning vulnerabilities. We propose R$^2$ECAP (Reused Reasoning-Cache Poisoning), a black-box attack that plants poisoned reasoning traces solely through interactions…

Read at arXiv (research papers)

Journal articles (OpenAlex) Weng Hong Tang · Analysis Geopolitics

Some Objections to Phenomenal Conservatism about Inferential Justification

Abstract Proponents of phenomenal conservatism appeal to one’s experiences – in particular, to one’s seemings or appearances – to account for non-inferential justification. Some extend this approach to the inferential case. On Michael Huemer’s view, inferential seemings yield inferential justification: if it seems that the conclusion of an inference is true in light of the premisses being true, then – provided our belief in the premisses is justified – we gain some…

Read at Journal articles (OpenAlex)

arXiv (research papers) Yichi Zhang SI

Secure Speculative Decoding for Large Language Models

Speculative decoding accelerates inference for a large language model (LLM), referred to as the \emph{target model}, by first using a smaller model, referred to as the \emph{draft model}, to generate candidate tokens and then verifying them with the target model for acceptance or rejection. Prior studies primarily focused on the efficiency-utility trade-off of speculative decoding, e.g., lossy speculative decoding, leaving its security implications largely unexplored. In…

Read at arXiv (research papers)

Journal articles (OpenAlex) Hamed Nozari, Zornitsa B. Yordanova · AgriEngineering

An Intelligent LLM–Fuzzy Decision Support and Digital Twin Framework for Adaptive Agricultural Robotics Systems

This study develops an integrated LLM–fuzzy decision-support and digital twin framework for adaptive agricultural robotics, combining semantic reasoning, fuzzy inference, multi-objective optimization, and feedback-driven state updating. Numerical sensor data, operational information, and textual field knowledge are transformed into validated decision rules and model-ready parameters, enabling coordinated decisions across economic, operational, crop-stress, and…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Zishuai Liu, Ruili Fang, Jin Lu et al. · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies Gadgets

LastAct: Trajectory-Guided Latest-Activity Localization for Real-Time Smart-Home Activity Recognition

Human Activity Recognition (HAR) from ambient sensors enables smart-home applications such as health monitoring and assisted living. In realistic deployments, however, sensor events arrive as a continuous stream and activity boundaries are unknown. Sliding-window inference therefore produces many windows that straddle transitions and contain mixed activities, creating boundary contamination that violates the pre-segmented instance assumption used by most benchmarks and…

Read at Journal articles (OpenAlex)

arXiv (research papers) Mark Russinovich SI

CounterSteer: Suppressing Indirect Prompt Injection with Activation Steering

Indirect prompt injection makes an LLM agent treat untrusted retrieved text as instructions. We present CounterSteer, an inference-time defense that suppresses this behavior inside the model. Per model, a five-step recipe fits a residual-stream direction from paired episodes differing only in whether an embedded instruction is followed, and retains it only if it passes pre-specified causal and capability gates. At deployment, the direction is subtracted from every…

Read at arXiv (research papers)

arXiv (research papers) Rohit Saxena SI

Nudgeability: Reasoning Models Follow Confidence Signals Without Tracking Their Own Competence

Reasoning language models that can call tools must decide during inference whether to answer unaided or delegate. Any self-reflection mechanism for this must answer three questions: where the reflective signal comes from (verbal reports, output distributions, hidden states, a separate predictor), how it is presented to the model (numerical prediction, confidence token, prompt injection), and whether it changes the model's subsequent action. We isolate the third question. At…

Read at arXiv (research papers)

Journal articles (OpenAlex) Andrew Franklin Johnson · Administrative Theory & Praxis Geopolitics

Politics everywhere: stigma and political perception in the contemporary United States

Political ideology has become a pervasive lens through which social and organizational settings are interpreted, a trend intensified through recent political developments. Across everyday interactions, consumer behaviors, and institutional settings, individuals rely on observable cues to infer others’ political beliefs. Though often rapid and inaccurate, these inferences are highly consequential as they function as stigma-based evaluations that attach moral and social…

Read at Journal articles (OpenAlex)

arXiv (research papers) Justin Szczepaniak SI

Defusing Explosive Prompts: Understanding and Preventing Trigger-Based Prompt Injections in LLM Agents

As LLM applications integrate with external tools, they are increasingly exposed to indirect prompt injection (IPI), where adversarial instructions are embedded in retrieved content. Conventional IPIs fire on contact: the moment an agent ingests the content, it carries out the instruction. We introduce the explosive prompt, a conditional payload that stays dormant until an attacker-chosen trigger is met, in effect a training-free, inference-time backdoor planted in a single…

Read at arXiv (research papers)

arXiv (research papers) Frank E. Bobe SI

Deep Noir: Autonomous Steering Discovery via Architectural Chronometry in Transformer Models

Activation steering modifies LLM behavior at inference time, but identifying where and how strongly to steer remains manual. We introduce Deep Noir, a framework that uses Logit Lens convergence and causal head-level attribution to autonomously discover optimal steering parameters. Across three scales (1B x 3, 2-3B x 2, and 7-9B x 4), our engine achieves 16.7 percentage-point improvement on spam at 1B (standard deviation 4.7; 39 runs), with gains increasing to 21 to 42…

Read at arXiv (research papers)

arXiv (research papers) Arslan Brömme SI

Intrusion Detection for Agentic Processes: Evidence-Based Runtime Monitoring

Agent deployments increasingly combine language-model inference with retrieval, delegation, tool execution, external-system access, and human approval. Security-relevant deviations can therefore emerge across an evolving process rather than in one isolated input or action. Building on the author's earlier black-box architecture for agentic processes and the subsequent evidence-claim model, this paper proposes an Agentic-Process Intrusion Detection System (A-IDS), an…

Read at arXiv (research papers)

arXiv (research papers) Lisa Bouger SI

Influence Score and Transformers interpretability: Measure of the Effective Impact of Attention Heads at inference time

We propose an influence score to quantify the contribution of attention heads to classification decisions in Transformer-based models designed for prompt injection detection. The score combines directional influence on the logits with structural contribution within the residual stream, enabling a multi-scale analysis at the head, layer, and network levels. Applied to a DeBERTa model specialized for prompt injection detection, our framework reveals distinct decision…

Read at arXiv (research papers)

Journal articles (OpenAlex) Andrew Choi, Daniel Choi, Junho Park · Proceedings of the Human Factors and Ergonomics Society Annual Meeting Gadgets

Confound-Aware Evaluation of Smart-Home Digital Biomarkers: Auditing Observation Intensity Dominance in Cognitive Status and Agitation

Confounding by observation intensity (“coverage”) is a validity risk in smart-home digital biomarkers: coverage can correlate with labels and dominate engineered activity features. We present a confound-aware evaluation protocol that compares coverage-only (Cov3), coverage++ (CovPP), and full-feature (All) models using participant-disjoint out-of-fold prediction, paired cluster-aware inference, and leakage audits. The protocol is demonstrated on CASAS scripted cognitive…

Read at Journal articles (OpenAlex)

arXiv (research papers) Jun He SI

Stored Is Not Supported: Typed Provenance and Assertion Guardrails for Persistent AI Agents

Persistent AI agents construct autobiographical state through reflection, retrieval, and consolidation. Persistence changes availability, not epistemic standing: stored or retrieved material is not thereby supported. Untrusted inputs, prompt injections, and model inferences can therefore enter persistent state and later be presented as agent history or user commitments. We specify typed provenance and assertion guardrails for autobiographical assertion boundedness, a…

Read at arXiv (research papers)

arXiv (research papers) Yichao Gao SI

What Guides the Agent? Adjudicating Unauthorized Behavior via Localizing Behavior-Guiding Instructions

LLM agents integrated with external resources gain complex task capabilities, yet the unified natural-language context channel makes them vulnerable to injection attacks: untrusted external data may be dynamically parsed as behavior-guiding instructions during LLM inference, thereby subverting the agent's decision. Existing defenses focus on static detection or isolation of malicious content at the input/output level, remains insufficient for detecting such dynamic…

Read at arXiv (research papers)

Journal articles (OpenAlex) Zhen Zhang, Xingyuan Hu · Informatica Space

Risk-Aware Task Allocation for LEO Satellite Constellations Using Deep Reinforcement Learning, Fuzzy Inference, and K-Means Clustering

The explosive expansion of Low Earth Orbit (LEO) satellite mega-constellations presents critical challenges in operational risk management, real-time task allocation, and dynamic resource management due to their inherent time-varying topology and systemic uncertainties. To overcome the limitations of traditional optimization methods, this paper proposes a novel, hybrid artificial intelligence framework integrating K-means Clustering, Fuzzy Logic (FL), and Deep Reinforcement…

Read at Journal articles (OpenAlex)

arXiv (research papers) Qianlong Yang SI

Mitigating Visual Degradation in MLLMs via Spatial-Spectral Visual Anchor Learning

Despite the progress of multimodal large language models (MLLMs), they continue to exhibit deficiencies in visual perception. Following visual instruction tuning, internal MLLM representations rapidly deviate from their original semantic states during inference, causing severe information degradation. While existing methods attempt to leverage external vision foundation models (VFMs) to align internal representations, we find that direct alignment with VFMs enhances visual…

Read at arXiv (research papers)

arXiv (research papers) Tejaswi V. Panchagnula SI

Inference-Time Mitigation of Adversarial Political Bias in Large Language Models

As Large Language Models (LLMs) become the mainstay for information retrieval and summarization tasks, ensuring that they are always non-partisan and invulnerable to political bias is a critical step towards safer and more trustworthy Artificial Intelligence (AI). Current model alignment paradigms, such as reinforcement learning from human feedback (RLHF), make LLMs follow overarching safety instructions. However, this instruction tuning can be exploited via adversarial…

Read at arXiv (research papers)

Berkeley AI Research blog SI

Intelligence is Free, Now What? Data Systems for, of, and by Agents

... government of the people, by the people, for the people ... — Abraham Lincoln, Gettysburg Address (1863) The cost of AI is dropping rapidly. GPT-4-class capabilities cost roughly $30 per million tokens in early 2023; today the same runs under $1, and some providers are pushing costs below $0.10. Across benchmarks, inference prices have fallen between 9x and 900x per year, with a median decline near 50x. Even frontier models are getting dramatically cheaper each…

Read at Berkeley AI Research blog

Simon Willison SI

Incident Report: CVE-2026-LGTM

Incident Report: CVE-2026-LGTM Spectacular hypothetical incident report by Andrew Nesbitt. Day 2, 16:00 UTC --- Two AI review agents from competing vendors, both attached to a downstream pull request bumping foxhole-lz4, enter a disagreement loop over whether the package is malicious. After 340 comments and $41,255 in inference spend, Finance revokes both API keys; one vendor's marketing team, cc'd on the cost anomaly alert, issues a press release citing "a 430% YoY increase…

Read at Simon Willison

arXiv (research papers) Xingwei Zhong SI

DE-FIVE: Detecting Malicious Image Prompts via Fourier Features and Image Vector Embeddings

Vision language models (VLMs) employ both visual and textual modalities to enable advanced vision-language inference. However, incorporating visual modalities expands the attack surface of VLMs, making them more susceptible to security threats such as adversarial perturbations and indirect prompt injection, wherein crafted malicious image prompts can elicit unintended model outputs. Existing defense methods against malicious image prompts remain insufficient as they…

Read at arXiv (research papers)

arXiv (research papers) Po-Han Cheng SI

CodeSentinel: A Three-Layer Defense Against Indirect Prompt Injection in Code Contexts

Code large language models increasingly retrieve external code context from repositories, documentation, issue threads, and coding-agent environments, creating an indirect prompt-injection surface where attackers hide instructions in comments, strings, identifiers, or decoy code. We propose CodeSentinel, a three-layer inference-time sanitizer. It uses Tree-sitter to extract high-risk model-facing CST nodes, then combines syntax-guided pre-filtering, CST-guided Dynamic…

Read at arXiv (research papers)

arXiv (research papers) Pengfei He SI

PI-Hunter: Automated Red-Teaming for Exposing and Localizing Prompt Injections

Large Language Models (LLMs) are rapidly evolving into agentic systems that interact with external tools and environments, introducing new security risks such as indirect prompt injection attacks through untrusted external sources. Existing defenses mainly focus on blocking malicious content at inference time, and current red-teaming methods primarily optimize attack success. As a result, developers have limited visibility into how latent prompt injections emerge and…

Read at arXiv (research papers)

Older stories