Search: “context”

Uncertainty-gated selection for block-sparse attention

model-releasesarxiv-cs-cl

arXiv:2607.07724v1 Announce Type: cross Abstract: Block-sparse attention scales long-context language models by replacing the O(N^2) softmax with a per-query top-k selection over key blocks. This cuto

10 Jul 2026

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

ALER-TI: Aligned Latent Embedding Retrieval for Time Series Imputation

DGX agent

arXiv:2607.07640v1 Announce Type: cross Abstract: Deep learning has significantly advanced time series imputation, yet most existing architectures primarily rely on localized temporal context within t

local-aiarxiv-cs-ai

9 Jul 2026

The Key to Going Linear: Analysis-Driven Transformer Linearization

arXiv:2607.07706v1 Announce Type: new Abstract: The quadratic cost of causal self-attention severely bottlenecks long-context transformer inference. While numerous post hoc linearization pipelines exi

9 Jul 2026

Understanding Interpretation Difficulty in Harmful Online Communication: Insights from Cybercrime Communities

arXiv:2607.07277v1 Announce Type: new Abstract: Harmful online communication often contains slang, coded terms, abbreviations, and community-specific expressions, which make messages difficult to inte

local-aiarxiv-cs-cl

9 Jul 2026

Hardware

Light-Omni: Reflex over Reasoning in Agentic Video Understanding with Long-Term Memory

arXiv:2607.05511v1 Announce Type: new Abstract: Agentic video understanding equips models with long-term memory to autonomously process and respond to continuous, long-horizon multimodal streams. Howe

hardwarearxiv-cs-cv

8 Jul 2026

Agent Data Injection Attacks are Realistic Threats to AI Agents

arXiv:2607.05120v1 Announce Type: cross Abstract: AI agents act on behalf of user prompts, consuming external data and taking actions based on the agent context. Prior research on AI agent security ha

Attending to Multimodal Generation One Token at a Time

arXiv:2607.03738v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) generate responses autoregressively, integrating visual and linguistic information in an evolving context. Pr

Detecting Hallucinations in Retrieval-Augmented Generation through Grounding-Aware Sensitivity by Perturbation (GASP)

arXiv:2607.04223v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) reduces but does not eliminate hallucination, and existing detectors return a single answer-level score that does

KVpop -- Key-Value Cache Compression with Predictive Online Pruning

arXiv:2607.05061v1 Announce Type: new Abstract: Key-value (KV) cache growth is a major bottleneck in autoregressive decoding, as memory and bandwidth scale linearly with context length. Existing KV ev

safetyarxiv-cs-lg

Language Models Represent and Transform Concepts with Shared Geometry

arXiv:2607.04525v1 Announce Type: cross Abstract: How concepts are represented in neural networks is a fundamental question in machine learning. The dominant view treats concept representations as sta

LLM-Guided Transportation Hub Capacity Planning with Textual Business Inputs

arXiv:2607.03651v1 Announce Type: new Abstract: While traditional hub capacity planning models optimize effectively for quantitative inputs, they often fail to digest qualitative business context. We

agentsarxiv-cs-lg

PreSIST: Vision-Language-Informed Object Persistence Prediction in Open-World Scenes

arXiv:2607.04057v1 Announce Type: new Abstract: Robots deployed over long periods must reason about environments that change over time. Existing long-term perception systems often address object chang

researcharxiv-cs-cv

Towards Standardized Light Field Quality Assessment: Hybrid Subjective Benchmarking and Objective Metric Evaluation

model-releasesarxiv-cs-cv

arXiv:2607.03494v1 Announce Type: new Abstract: Benchmarking immersive media coding solutions, especially in the standardization context, requires reliable and reproducible subjective quality assessme

NarrativeTrack: Evaluating Entity-Centric Reasoning for Narrative Understanding

arXiv:2601.01095v3 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) have achieved impressive progress in vision-language reasoning, yet their ability to understand tempo

3 Jul 2026

An Empirical Study of Security Calibration in Large Language Models for Code

arXiv:2606.31159v1 Announce Type: cross Abstract: Large Language Models (LLMs) are rapidly transforming software development, yet their use in security-critical contexts raises a key question: do mode

1 Jul 2026

Disentangling Reasoning Logic to Resolve Explicit Knowledge Conflicts

arXiv:2508.01273v3 Announce Type: replace Abstract: Explicit knowledge conflicts, occurring when retrieved contexts contain contradictory information, pose a fundamental challenge for Large Language M

1 Jul 2026

Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models

arXiv:2606.31603v1 Announce Type: cross Abstract: Semantic segmentation models struggle with data sparsity and rare or visually diverse regions, e.g., dense regions or small objects in aerial or auton

agentsarxiv-cs-ai

1 Jul 2026

Can LLMs Reliably Self-Report Adversarial Prefills, and How?

arXiv:2606.23671v2 Announce Type: replace Abstract: Prior work shows that large language models (LLMs) exhibit introspective capability on benign tasks. We extend the question to safety contexts and e

safetyarxiv-cs-cl

CaveAgent: Transforming LLMs into Stateful Runtime Operators

arXiv:2601.01569v4 Announce Type: replace Abstract: LLM-based agents are increasingly capable of complex task execution, yet current agentic systems remain constrained by text-centric paradigms that s

agentsarxiv-cs-ai

Coverage-Driven KV Cache Eviction for Efficient and Improved Inference of LLM

arXiv:2606.29563v1 Announce Type: cross Abstract: Large language models (LLMs) excel at complex tasks like question answering and summarization, thanks to their ability to handle long-context inputs.

Defeat Devices in AI Systems

arXiv:2606.28863v1 Announce Type: cross Abstract: AI systems increasingly exhibit behavior that differs systematically between evaluation and deployment contexts. Alignment faking, sandbagging, benchm

Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions

arXiv:2507.05257v4 Announce Type: replace-cross Abstract: Recent benchmarks for Large Language Model (LLM) agents primarily focus on evaluating reasoning, planning, and execution capabilities, while a

Know Before You Fetch: Calibrated Retrieval-Budget Allocation for Retrieval-Augmented Generation

model-releasesarxiv-cs-cl

arXiv:2606.29959v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) typically retrieves a fixed number of passages for every query. This is wasteful when the reader already knows th

LEDGER: Scaling Agentic Document Editing with Dependency-aware Graph Retrieval

arXiv:2606.28379v1 Announce Type: cross Abstract: We introduce LEDGER to tackle the novel context engineering challenge of agentic document editing, where localized edits to long, structured documents

Applications

Schema-First Retrieval: Embedding Catalogs for Natural Language Analytics

arXiv:2606.28387v1 Announce Type: cross Abstract: Enterprise text-to-SQL systems often fail before SQL is generated: the model receives the wrong schema context. Modern warehouses contain thousands of

applicationsarxiv-cs-ai

Sparse Autoencoders are Capable LLM Jailbreak Mitigators

arXiv:2602.12418v2 Announce Type: replace-cross Abstract: Jailbreak attacks remain a persistent threat to large language model safety. We propose Context-Conditioned Delta Steering (CC-Delta), an SAE-

safetyarxiv-cs-cl

Tutorials

SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents

arXiv:2606.28434v1 Announce Type: cross Abstract: Long-horizon software engineering agents often need to manage lengthy and noisy interaction histories under limited context budgets. Existing memory m

tutorialsarxiv-cs-ai

HiMu: Hierarchical Multimodal Frame Selection for Long Video Question Answering

arXiv:2603.18558v2 Announce Type: replace-cross Abstract: Long-form video question answering requires reasoning over extended temporal contexts, making frame selection a critical bottleneck for multi-

Mitigating Position Bias in Transformers via Layer-Specific Positional Embedding Scaling

arXiv:2606.27705v1 Announce Type: new Abstract: Large Language Models (LLMs) still struggle with the ``lost-in-the-middle'' problem, where critical information located in the middle of long-context in

safetyarxiv-cs-cl

NLL-Guided Full-Attention Layer Selection for Training-Free Sliding-Window Adaptation

arXiv:2606.27791v1 Announce Type: cross Abstract: Hybrid attention models that mix full and sliding-window attention across layers offer a promising approach to efficient long-context inference, but t

TempAct: Advancing Temporal Plausibility in Autoregressive Video Generation via Planner-Executor RL

arXiv:2606.28016v1 Announce Type: new Abstract: Autoregressive (AR) video diffusion models enable low-latency streaming generation by synthesizing videos chunk by chunk with cached visual context, but

safetyarxiv-cs-cv

When Multi-Robot Systems Meet Agentic AI:Towards Embodied Collective Intelligence

arXiv:2606.27929v1 Announce Type: new Abstract: Embodied AI is increasingly becoming agentic, shifting robots from perception--control pipelines towards closed-loop systems that can retrieve context,

agentsarxiv-cs-ro

Application of LLMs to Threat Assessment of Foreign Peacekeeping Missions

arXiv:2606.27106v1 Announce Type: cross Abstract: We present a novel approach for applying Large Language Models (LLMs) to threat assessment in the context of foreign peacekeeping missions. Building o

Information-Aware KV Cache Compression for Long Reasoning

arXiv:2606.26875v1 Announce Type: cross Abstract: Reasoning capability has advanced rapidly in large language models (LLMs), leading to an increasing size of key-value (KV) cache in both prefilling an

Speaking Numbers to LLMs: Multi-Wavelet Number Embeddings for Time Series Forecasting

arXiv:2606.26487v1 Announce Type: cross Abstract: Large language models (LLMs) are attractive for context-aware time series forecasting because they can integrate heterogeneous textual signals, yet th

local-aiarxiv-cs-ai

The Spec Growth Engine: Spec-Anchored, Code-Coupled, Drift-Enforced Architecture for AI-Assisted Software Development

arXiv:2606.27045v1 Announce Type: cross Abstract: AI coding agents dramatically accelerate implementation speed but introduce two structural failure modes that existing spec-driven approaches do not f

Towards Explainable Adjudicative Variance: Quantifying Judicial Discretion via Gated Multi-Task Learning

model-releasesarxiv-cs-cl

arXiv:2606.27069v1 Announce Type: new Abstract: Legal outcome prediction must disentangle objective case facts from adjudicative context. Merit-based rulings rely on factual evidence while technical d

ATMA: Length-Invariant Language Modeling via Polar Attention and Gated-Delta Compression Memory

arXiv:2606.25156v1 Announce Type: new Abstract: Modern large language models based on softmax scaled-dot-product attention are constrained by their training sequence length: as the key-value sequence

local-aiarxiv-cs-lg

Attractive and Repulsive Pattern Control in Sequence Generation

arXiv:2606.24911v1 Announce Type: cross Abstract: Variable-order Markov models preserve local symbolic syntax by adapting context length, but long continuations can enter recurring high-order 'tunnels

Privacy Vulnerabilities of Attention Layers in Tabular Foundation Models and Protection of High-Risk Queries

arXiv:2606.26021v1 Announce Type: cross Abstract: Tabular foundation models are commonly assumed to present limited privacy concerns as they are often pre-trained on large collections of synthetic dat

Why Do Accumulated Transformations Extrapolate?

arXiv:2606.24975v1 Announce Type: cross Abstract: PaTH Attention showed that replacing RoPE's position-indexed rotations with accumulated data-dependent Householder reflections yields strong length ex

researcharxiv-cs-cl

Same Lesson, Different Story: Cross-Lingual Reconstruction of Cultural Narratives in Large Language Models

arXiv:2606.24610v1 Announce Type: new Abstract: The evaluation of cultural grounding context becomes complex when multiple cultures convey the same moral lesson. This challenge is particularly relevan

researcharxiv-cs-cl

24 Jun 2026

When Retrieval Metrics Mislead: Measuring Policy Signal in Long-Horizon Tool-Use Agents

arXiv:2606.23937v1 Announce Type: cross Abstract: Exact-match retrieval recall is often used as a proxy for whether a retriever supplies useful policy context to a downstream decision model. We test t

24 Jun 2026

Continuous Behavioral Authentication via Multi-Expert BERT Log Analysis for Secure Data Sharing

arXiv:2606.21900v1 Announce Type: cross Abstract: Continuous authentication for mobile and zero-trust systems requires nonintrusive evidence confirming the enrolled user-device context remains valid a

safetyarxiv-cs-lg

Does RoPE Prevent or Degrade Retrieval Heads? A Mechanistic Analysis Across Model Families

arXiv:2606.21249v1 Announce Type: new Abstract: Retrieval heads, attention heads that copy information from earlier context to the current position, have been proposed as the mechanistic substrate for

Formalizing Task-Space Complexity for Zero-Shot Generalization

arXiv:2606.20967v1 Announce Type: new Abstract: Policies must operate across diverse conditions, yet a single policy is often conservative while fully adaptive schemes can be complex. We study zero-sh

safetyarxiv-cs-lg

Hardware

HERALD: High-Throughput Block Diffusion LLM Serving via CPU-GPU Cooperative KV Cache Retrieval

arXiv:2606.21633v1 Announce Type: new Abstract: Diffusion LLMs (dLLMs) improve GPU utilization over autoregressive decoding by generating multiple tokens per forward pass, but their KV cache still gro

hardwarearxiv-cs-lg

Kamera: Unified Position-Invariant Multimodal KV Cache for Training-Free Reuse

model-releasesarxiv-cs-cv

arXiv:2606.23581v1 Announce Type: cross Abstract: Multimodal agents repeatedly re-examine the same video frames, UI screenshots, and rendered artifacts as their context window slides and reasoning ite