arXiv Machine Learning
Jul 3
Embedding Inference Attack
arXiv:2607. 01276v1 Announce Type: cross Abstract: Embedding models are essential components of modern Information Retrieval (IR) systems, yet they are typically hidden behind APIs.
Large language models
Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv AI
Jul 3
DemoPSD: Disagreement-Modulated Policy Self-Distillation
arXiv:2607. 02502v1 Announce Type: cross Abstract: On-policy self-distillation (OPSD) has emerged as a practical method for training large language models (LLMs) to reason, where a single model acts as both the teacher and the student with different levels of information access.
arXiv AI
Jul 3
An Isotropic Approach to Efficient Uncertainty Quantification with Gradient Norms
arXiv:2603. 29466v2 Announce Type: replace-cross Abstract: Existing methods for quantifying predictive uncertainty in neural networks are either computationally intractable for large language models or require access to training data that is typically unavailable.
arXiv AI
Jul 3
Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts
arXiv:2607. 01767v1 Announce Type: new Abstract: As agent planning moves from short tool chains toward persistent workflows with thousands or tens of thousands of steps, failures will occur inside large planning graphs rather than in isolated predictions.
arXiv AI
Jul 3
DRIFTLENS: Measuring Memory-Induced Reasoning Drift in Personalized Language Models
arXiv:2607. 02374v1 Announce Type: new Abstract: Personalization changes what a model says to a user; we show that it can also change the reasoning trajectory used to justify the response.
arXiv Machine Learning
Jul 3
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 temporally unfolding narratives in videos remains underexplored.
arXiv AI
Jul 3
Activation Steering for Aligned Open-ended Generation without Sacrificing Coherence
arXiv:2604. 08169v2 Announce Type: replace Abstract: Alignment in LLMs is more brittle than commonly assumed: misalignment can be induced by adversarial prompts, benign fine-tuning, emergent misalignment, and goal misgeneralization.
arXiv AI
Jul 3
Towards Load-Aware Prefill Deflection for Disaggregated LLM Serving
arXiv:2607. 02043v1 Announce Type: cross Abstract: Disaggregated LLM serving runs prefill and decode on separate GPU pools to keep the two phases from interfering.
arXiv Machine Learning
Jul 3
DeadPool: Resilient LLM Training with Hot-Swapping via Zero-Overhead Checkpoint
arXiv:2607. 01646v1 Announce Type: new Abstract: State-of-the-art large language model (LLM) training takes tens of thousands of graphics processing units (GPUs) for months and encounters failures across the software and hardware stack.
arXiv AI
Jul 3
GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation
arXiv:2606. 22737v2 Announce Type: replace Abstract: Before letting an agent operate over real context, can you prove it used the right evidence?
arXiv AI
Jul 3
SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use
arXiv:2607. 01874v1 Announce Type: new Abstract: Skills are becoming a reusable operational layer for LLM agents, encoding SOPs, domain rules, tool workflows, scripts, and validation routines.
arXiv AI
Jul 3
LACUNA: A Testbed for Evaluating Localization Precision for LLM Unlearning
arXiv:2607. 02513v1 Announce Type: cross Abstract: LLMs memorize sensitive training data, including personally identifiable information (PII), creating a pressing need for reliable post hoc removal methods.
arXiv AI
Jul 3
CreativityPrism: A Cross-Domain Evaluation Framework for Large Language Model Creativity
arXiv:2510. 20091v3 Announce Type: replace-cross Abstract: Creativity is often seen as a hallmark of human intelligence.
arXiv AI
Jul 3
Multilingual Prompt Localization for Agent-as-a-Judge: Language and Backbone Sensitivity in Requirement-Level Evaluation
arXiv:2604. 04532v2 Announce Type: replace-cross Abstract: Evaluation language is typically treated as a fixed English default in agentic code benchmarks, yet we show that changing the judge's language can invert backbone rankings.
arXiv AI
Jul 3
Scaling Trends for Lie Detector Oversight in Preference Learning
arXiv:2607. 01567v1 Announce Type: new Abstract: Deceptive behavior in LLMs is costly to monitor and prevent, motivating approaches such as Scalable Oversight via Lie Detectors (SOLiD) (Cundy & Gleave, 2025), which uses lie detectors to identify responses for review by high-cost labelers.
arXiv AI
Jul 3
MetaTT: A Global Tensor-Train Adapter for Parameter-Efficient Fine-Tuning
arXiv:2506. 09105v3 Announce Type: replace-cross Abstract: We present MetaTT, a Tensor Train (TT) adapter framework for fine-tuning of pre-trained transformers.
arXiv Machine Learning
Jul 3
FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation
arXiv:2603. 09046v3 Announce Type: replace-cross Abstract: Device-side Large Language Models (LLMs) have witnessed explosive growth, offering higher privacy and availability compared to cloud-side LLMs.
arXiv AI
Jul 3
TokenScope: Token-Level Explainability and Interpretability for Code-Oriented Tasks in Large Language Models
arXiv:2607. 01235v1 Announce Type: cross Abstract: Understanding how Large Language Models (LLMs) make token-level decisions during code generation remains a major challenge for both researchers and practitioners.
arXiv AI
Jul 3
CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation
arXiv:2603. 22435v2 Announce Type: replace-cross Abstract: "Code-as-Policy" considers how executable code can complement data-intensive Vision-Language-Action (VLA) methods, yet their effectiveness as autonomous controllers for embodied manipulation remains underexplored.
arXiv Machine Learning
Jul 3
FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation
arXiv:2606. 23370v2 Announce Type: replace-cross Abstract: Device-side Large Language Models (LLMs) have grown explosively, offering stronger privacy and higher availability than their cloud-side counterparts.