CUEBES

Farewell to Item IDs: Unlocking the Scaling Potential of Large Ranking Models via Semantic Tokens

arXiv:2601.22694v1 Announce Type: new Abstract: Recent studies on scaling up ranking models have achieved substantial improvement for recommendation systems and search engines. However, most large-scale...

Software Neuroscience

arXiv CS Feb 2

Bi-MCQ: Reformulating Vision-Language Alignment for Negation Understanding

arXiv:2601.22696v1 Announce Type: new Abstract: Recent vision-language models (VLMs) achieve strong zero-shot performance via large-scale image-text pretraining and have been widely adopted in medical image...

Medicine & Health Policy

arXiv CS Feb 2

Models Know Models Best: Evaluation via Model-Preferred Formats

arXiv:2601.22699v1 Announce Type: new Abstract: Performance of Large Language Models (LLMs) on multiple-choice tasks differs markedly between symbol-based and cloze-style evaluation formats. The observed discrepancies...

Software Politics

arXiv CS Feb 2

Best-of-Q: Improving VLM agents with Q-function Action Ranking at Inference

arXiv:2601.22701v1 Announce Type: new Abstract: Vision-Language Models (VLMs) have become powerful backbones for agents to autonomously operate in digital environments like the web and operating...

Policy Robotics

arXiv CS Feb 2

DAVIS: OOD Detection via Dominant Activations and Variance for Increased Separation

arXiv:2601.22703v1 Announce Type: new Abstract: Detecting out-of-distribution (OOD) inputs is a critical safeguard for deploying machine learning models in the real world. However, most post-hoc...

World News Technology

arXiv CS Feb 2

Multi-target DoA estimation with a single Rydberg atomic receiver by spectral analysis of spatially-resolved fluorescence

arXiv:2601.22704v1 Announce Type: new Abstract: Rydberg-based Direction-of-Arrival (DoA) estimation has been hampered by the complexity of receiver arrays and the single-target, narrow-band limitations of existing...

Quantum Computing Software

arXiv CS Feb 2

CONCUR: High-Throughput Agentic Batch Inference of LLM via Congestion-Based Concurrency Control

arXiv:2601.22705v1 Announce Type: new Abstract: Batch inference for agentic workloads stresses the GPU key-value (KV) cache in a sustained and cumulative manner, often causing severe...

Policy Artificial Intelligence

arXiv CS Feb 2

RealSec-bench: A Benchmark for Evaluating Secure Code Generation in Real-World Repositories

arXiv:2601.22706v1 Announce Type: new Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities in code generation, but their proficiency in producing secure code remains a...

Technology Software

arXiv CS Feb 2

Deep Learning-Based Early-Stage IR-Drop Estimation via CNN Surrogate Modeling

arXiv:2601.22707v1 Announce Type: new Abstract: IR-drop is a critical power integrity challenge in modern VLSI designs that can cause timing degradation, reliability issues, and functional...

Software Biology

arXiv CS Feb 2

A Unified Study of LoRA Variants: Taxonomy, Review, Codebase, and Empirical Evaluation

arXiv:2601.22708v1 Announce Type: new Abstract: Low-Rank Adaptation (LoRA) is a fundamental parameter-efficient fine-tuning method that balances efficiency and performance in large-scale neural networks. However, the...

Biology Neuroscience

arXiv CS Feb 2

Gated Relational Alignment via Confidence-based Distillation for Efficient VLMs

arXiv:2601.22709v1 Announce Type: new Abstract: Vision-Language Models (VLMs) achieve strong multimodal performance but are costly to deploy, and post-training quantization often causes significant accuracy loss....

Policy

arXiv CS Feb 2

AlienLM: Alienization of Language for API-Boundary Privacy in Black-Box LLMs

arXiv:2601.22710v1 Announce Type: new Abstract: Modern LLMs are increasingly accessed via black-box APIs, requiring users to transmit sensitive prompts, outputs, and fine-tuning data to external...

Cybersecurity Mathematics

arXiv CS Feb 2

SQUAD: Scalable Quorum Adaptive Decisions via ensemble of early exit neural networks

arXiv:2601.22711v1 Announce Type: new Abstract: Early-exit neural networks have become popular for reducing inference latency by allowing intermediate predictions when sufficient confidence is achieved. However,...

Technology Neuroscience

arXiv CS Feb 2

Vision-Language Models Unlock Task-Centric Latent Actions

arXiv:2601.22714v1 Announce Type: new Abstract: Latent Action Models (LAMs) have rapidly gained traction as an important component in the pre-training pipelines of leading Vision-Language-Action models....

World News Policy

arXiv CS Feb 2

Breaking the Blocks: Continuous Low-Rank Decomposed Scaling for Unified LLM Quantization and Adaptation

arXiv:2601.22716v1 Announce Type: new Abstract: Current quantization methods for LLMs predominantly rely on block-wise structures to maintain efficiency, often at the cost of representational flexibility....

Software Energy

arXiv CS Feb 2

A Step Back: Prefix Importance Ratio Stabilizes Policy Optimization

arXiv:2601.22718v1 Announce Type: new Abstract: Reinforcement learning (RL) post-training has increasingly demonstrated strong ability to elicit reasoning behaviors in large language models (LLMs). For training...

Policy Psychology

arXiv CS Feb 2

AEGIS: White-Box Attack Path Generation using LLMs and Training Effectiveness Evaluation for Large-Scale Cyber Defence Exercises

arXiv:2601.22720v1 Announce Type: new Abstract: Creating attack paths for cyber defence exercises requires substantial expert effort. Existing automation requires vulnerability graphs or exploit sets curated...

Robotics Technology

arXiv CS Feb 2

Local Intrinsic Dimension of Representations Predicts Alignment and Generalization in AI Models and Human Brain

arXiv:2601.22722v1 Announce Type: new Abstract: Recent work has found that neural networks with stronger generalization tend to exhibit higher representational alignment with one another across...

Neuroscience Artificial Intelligence

arXiv CS Feb 2

OpenVTON-Bench: A Large-Scale High-Resolution Benchmark for Controllable Virtual Try-On Evaluation

arXiv:2601.22725v1 Announce Type: new Abstract: Recent advances in diffusion models have significantly elevated the visual fidelity of Virtual Try-On (VTON) systems, yet reliable evaluation remains...

Energy Psychology

arXiv CS Feb 2

On Small Pair Decompositions for Point Sets

arXiv:2601.22728v1 Announce Type: new Abstract: $\newcommand{\Re}{\mathbb{R}}$We study the minWSPD problem of computing the minimum-size well-separated pairs decomposition of a set of points, and show constant...

Software Policy

arXiv CS Feb 2

GaussianOcc3D: A Gaussian-Based Adaptive Multi-modal 3D Occupancy Prediction

arXiv:2601.22729v1 Announce Type: new Abstract: 3D semantic occupancy prediction is a pivotal task in autonomous driving, providing a dense and fine-grained understanding of the surrounding...

Biology Robotics

arXiv CS Feb 2

ImgCoT: Compressing Long Chain of Thought into Compact Visual Tokens for Efficient Reasoning of Large Language Model

arXiv:2601.22730v1 Announce Type: new Abstract: Compressing long chains of thought (CoT) into compact latent tokens is crucial for efficient reasoning with large language models (LLMs)....

Software World News

arXiv CS Feb 2

MM-THEBench: Do Reasoning MLLMs Think Reasonably?

arXiv:2601.22735v1 Announce Type: new Abstract: Recent advances in multimodal large language models (MLLMs) mark a shift from non-thinking models to post-trained reasoning models capable of...

Neuroscience Policy

arXiv CS Feb 2

Decomposing Epistemic Uncertainty for Causal Decision Making

arXiv:2601.22736v1 Announce Type: new Abstract: Causal inference from observational data provides strong evidence for the best action in decision-making without performing expensive randomized trials. The...

Artificial Intelligence Neuroscience