arXiv AI
Jul 29
Similar Models Learn Differently: Final-Window Pretraining Shapes Post-Training Beyond SFT
arXiv:2607. 25063v1 Announce Type: new Abstract: Developers judge a model checkpoint by how it behaves.
AI safety and alignment
Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv AI
Jul 29
Deep Delta Learning
arXiv:2601. 00417v4 Announce Type: replace-cross Abstract: Transformer residual streams evolve through additive updates.
arXiv AI
Jul 29
The Scaling Properties of Implicit Deductive Reasoning in Transformers
arXiv:2605. 04330v2 Announce Type: replace Abstract: We investigate the scaling properties of implicit deductive reasoning over Horn clauses in depth-bounded Transformers.
arXiv AI
Jul 29
CARE-MH: Towards Unified, Reproducible, and Comparable Evaluation of Mental Health LLMs
arXiv:2607. 24754v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used to provide mental health support, requiring reliable evaluation of safety, empathy, and therapeutic appropriateness.
arXiv AI
Jul 29
DDSNet: Dual-domain Symmetry-aware Network for PCSEL Property Prediction
arXiv:2607. 24785v1 Announce Type: cross Abstract: Efficient exploration of the photonic crystal (PhC) lattice design space is essential for developing photonic crystal surface-emitting lasers.
arXiv AI
Jul 29
Explanation-Bound Tool Execution for AI Agents: Server-Verified Action Claims Without Trusting Model Rationales
arXiv:2607. 25364v1 Announce Type: new Abstract: Tool-using agents expose structured calls but commonly attach free-form rationales.
arXiv AI
Jul 29
The Possibility of Artificial Intelligence Becoming a Subject and the Alignment Problem
arXiv:2604. 14990v2 Announce Type: replace Abstract: The prospect of Artificial General Intelligence (AGI) is increasingly driving institutional decisions, and alignment of AGI is a hard problem.
arXiv AI
Jul 29
Towards Understanding the Cognitive Habits of Large Reasoning Models
arXiv:2506. 21571v3 Announce Type: replace-cross Abstract: Large Reasoning Models (LRMs), which autonomously produce a reasoning Chain of Thought (CoT) before producing final responses, offer a promising approach to interpreting and monitoring model behaviors.
arXiv AI
Jul 29
I2VShield: An Efficient Proactive Defense Framework against DiT-based Image-to-Video Models
arXiv:2607. 25522v1 Announce Type: cross Abstract: The rapid advancement of video generation models has led to the increasing misuse of image-to-video (I2V) models.
arXiv AI
Jul 29
Mechanisms of Width Scaling in Normalized Residual Networks: The Effective Alignment Dimension
arXiv:2607. 24887v1 Announce Type: cross Abstract: Existing theories of neural-network width characterize asymptotic limits, but provide limited guidance on whether an expansion direction identified from finite training data remains beneficial on unseen data.
arXiv AI
Jul 29
Instruction-Tuned Language Models Cannot Sample from Distributions They Can Describe
arXiv:2607. 25292v1 Announce Type: new Abstract: Silicon sampling uses language models as proxies for human survey respondents, treating each model call as an independent draw from the persona's response distribution.
arXiv AI
Jul 29
CIFNet: An Analytic Neural Learning Framework for Efficient and Calibrated Class-Incremental Learning
arXiv:2509. 11285v2 Announce Type: replace-cross Abstract: Class-Incremental Learning (CIL) in deep neural networks is conventionally framed as an iterative gradient-based optimization problem, incurring high computational cost, hyperparameter sensitivity, and risk of catastrophic forgetting.
arXiv AI
Jul 29
Quotient Dynamics, Effective Curvature, and Implicit Bias in Positive Quadratic Networks
arXiv:2607. 25624v1 Announce Type: new Abstract: Positive quadratic networks admit the low-rank representation f_U(x)=x^top UU^top x, where Uinmathbb{R}^{dtimes r} is identifiable only up to right orthogonal multiplication, representing a rank-r PSD matrix Q=UU^top.
arXiv AI
Jul 29
The LAIA Dataset: Labelled Attention for Intelligent Automobiles
arXiv:2607. 25570v1 Announce Type: cross Abstract: The development of autonomous vehicles (AVs) usually relies heavily on data-driven artificial intelligence (AI) models that require large volumes of sensor data with ground-truth annotations.
arXiv AI
Jul 29
HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising
arXiv:2607. 24779v1 Announce Type: new Abstract: Online advertising bidding systems typically deploy multiple offline-trained expert models (e.
arXiv AI
Jul 29
A Cost-Effective Multimodal LLM Reasoning Framework for Question Answering over Irregular Clinical Time Series
arXiv:2607. 25947v1 Announce Type: new Abstract: Question answering (QA) over irregular clinical time series (ICTS) plays a pivotal role in a wide range of healthcare applications.
arXiv AI
Jul 29
PatchWorld: Gradient-Free Optimization of Executable World Models for Agent Environments
arXiv:2605. 30880v4 Announce Type: replace-cross Abstract: World models for interactive text agents must typically be learned from observation-action trajectories alone.
arXiv AI
Jul 29
IRIS: Reusable Identity Representations from Frozen LLMs for Entity Alignment
arXiv:2607. 25579v1 Announce Type: cross Abstract: Entity alignment (EA) identifies entities across knowledge graphs (KGs) that refer to the same real-world object.
arXiv AI
Jul 29
Architectural Backdoors in Vision-Language Model Supply Chains via Representation Steering
arXiv:2607. 25479v1 Announce Type: cross Abstract: Vision--Language Models (VLMs) are increasingly deployed through a model supply chain in which pretrained checkpoints, architecture definitions, text encoders, and exported computation graphs are distributed by third parties and reused across downstream services.
arXiv AI
Jul 29
COVENANT: Natural-Language Workflow Compilation for Aligned Agent Execution
arXiv:2607. 25400v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly entrusted with natural-language workflow instructions (e.