One-Eval: An Agentic System for Automated and Traceable LLM Evaluation
arXiv:2603.09821v1 Announce Type: new Abstract: Reliable evaluation is essential for developing and deploying large language models, yet in practice it often requires substantial manual effort: practitioners must identify appropriate benchmarks, reproduce heterogeneous evaluation codebases, configure dataset schema mappings, and interpret...
Model Merging in the Era of Large Language Models: Methods, Applications, and Future Directions
arXiv:2603.09938v1 Announce Type: new Abstract: Model merging has emerged as a transformative paradigm for combining the capabilities of multiple neural networks into a single unified model without additional training. With the rapid proliferation of fine-tuned large language models~(LLMs), merging techniques...
Self-hosted Lecture-to-Quiz: Local LLM MCQ Generation with Deterministic Quality Control
arXiv:2603.08729v1 Announce Type: cross Abstract: We present an end-to-end self-hosted (API-free) pipeline, where API-free means that lecture content is not sent to any external LLM service, that converts lecture PDFs into multiple-choice questions (MCQs) using a local LLM plus deterministic...
BiCLIP: Domain Canonicalization via Structured Geometric Transformation
arXiv:2603.08942v1 Announce Type: cross Abstract: Recent advances in vision-language models (VLMs) have demonstrated remarkable zero-shot capabilities, yet adapting these models to specialized domains remains a significant challenge. Building on recent theoretical insights suggesting that independently trained VLMs are related by...
A New Modeling to Feature Selection Based on the Fuzzy Rough Set Theory in Normal and Optimistic States on Hybrid Information Systems
arXiv:2603.08900v1 Announce Type: new Abstract: Considering the high volume, wide variety, and rapid speed of data generation, investigating feature selection methods for big data presents various applications and advantages. By removing irrelevant and redundant features, feature selection reduces data dimensions,...
Quantifying Memorization and Privacy Risks in Genomic Language Models
arXiv:2603.08913v1 Announce Type: new Abstract: Genomic language models (GLMs) have emerged as powerful tools for learning representations of DNA sequences, enabling advances in variant prediction, regulatory element identification, and cross-task transfer learning. However, as these models are increasingly trained or...
The $qs$ Inequality: Quantifying the Double Penalty of Mixture-of-Experts at Inference
arXiv:2603.08960v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models deliver high quality at low training FLOPs, but this efficiency often vanishes at inference. We identify a double penalty that structurally disadvantages MoE architectures during decoding: first, expert routing fragments microbatches and...
Semantic Level of Detail: Multi-Scale Knowledge Representation via Heat Kernel Diffusion on Hyperbolic Manifolds
arXiv:2603.08965v1 Announce Type: new Abstract: AI memory systems increasingly organize knowledge into graph structures -- knowledge graphs, entity relations, community hierarchies -- yet lack a principled mechanism for continuous resolution control: where do the qualitative boundaries between abstraction levels lie,...
MAPLE: Elevating Medical Reasoning from Statistical Consensus to Process-Led Alignment
arXiv:2603.08987v1 Announce Type: new Abstract: Recent advances in medical large language models have explored Test-Time Reinforcement Learning (TTRL) to enhance reasoning. However, standard TTRL often relies on majority voting (MV) as a heuristic supervision signal, which can be unreliable in...
SCALAR: Learning and Composing Skills through LLM Guided Symbolic Planning and Deep RL Grounding
arXiv:2603.09036v1 Announce Type: new Abstract: LM-based agents excel when given high-level action APIs but struggle to ground language into low-level control. Prior work has LLMs generate skills or reward functions for RL, but these one-shot approaches lack feedback to correct...
Sim2Act: Robust Simulation-to-Decision Learning via Adversarial Calibration and Group-Relative Perturbation
arXiv:2603.09053v1 Announce Type: new Abstract: Simulation-to-decision learning enables safe policy training in digital environments without risking real-world deployment, and has become essential in mission-critical domains such as supply chains and industrial systems. However, simulators learned from noisy or biased real-world...
Overcoming Valid Action Suppression in Unmasked Policy Gradient Algorithms
arXiv:2603.09090v1 Announce Type: new Abstract: In reinforcement learning environments with state-dependent action validity, action masking consistently outperforms penalty-based handling of invalid actions, yet existing theory only shows that masking preserves the policy gradient theorem. We identify a distinct failure mode...
Wrong Code, Right Structure: Learning Netlist Representations from Imperfect LLM-Generated RTL
arXiv:2603.09161v1 Announce Type: new Abstract: Learning effective netlist representations is fundamentally constrained by the scarcity of labeled datasets, as real designs are protected by Intellectual Property (IP) and costly to annotate. Existing work therefore focuses on small-scale circuits with clean...
Democratising Clinical AI through Dataset Condensation for Classical Clinical Models
arXiv:2603.09356v1 Announce Type: new Abstract: Dataset condensation (DC) learns a compact synthetic dataset that enables models to match the performance of full-data training, prioritising utility over distributional fidelity. While typically explored for computational efficiency, DC also holds promise for healthcare...
Zoom introduces an AI-powered office suite, says AI avatars for meetings arrive this month
Zoom is also introducing real-time deepfake detection tech for meetings.
"Dark Triad" Model Organisms of Misalignment: Narrow Fine-Tuning Mirrors Human Antisocial Behavior
arXiv:2603.06816v1 Announce Type: new Abstract: The alignment problem refers to concerns regarding powerful intelligences, ensuring compatibility with human preferences and values as capabilities increase. Current large language models (LLMs) show misaligned behaviors, such as strategic deception, manipulation, and reward-seeking, that...
A Dynamic Self-Evolving Extraction System
arXiv:2603.06915v1 Announce Type: new Abstract: The extraction of structured information from raw text is a fundamental component of many NLP applications, including document retrieval, ranking, and relevance estimation. High-quality extractions often require domain-specific accuracy, up-to-date understanding of specialized taxonomies, and...
A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness
arXiv:2603.06594v1 Announce Type: new Abstract: Automated \enquote{LLM-as-a-Judge} frameworks have become the de facto standard for scalable evaluation across natural language processing. For instance, in safety evaluation, these judges are relied upon to evaluate harmfulness in order to benchmark the robustness...
Can Safety Emerge from Weak Supervision? A Systematic Analysis of Small Language Models
arXiv:2603.07017v1 Announce Type: new Abstract: Safety alignment is critical for deploying large language models (LLMs) in real-world applications, yet most existing approaches rely on large human-annotated datasets and static red-teaming benchmarks that are costly, difficult to scale, and slow to...
AutoChecklist: Composable Pipelines for Checklist Generation and Scoring with LLM-as-a-Judge
arXiv:2603.07019v1 Announce Type: new Abstract: Checklists have emerged as a popular approach for interpretable and fine-grained evaluation, particularly with LLM-as-a-Judge. Beyond evaluation, these structured criteria can serve as signals for model alignment, reinforcement learning, and self-correction. To support these use...
Lying to Win: Assessing LLM Deception through Human-AI Games and Parallel-World Probing
arXiv:2603.07202v1 Announce Type: new Abstract: As Large Language Models (LLMs) transition into autonomous agentic roles, the risk of deception-defined behaviorally as the systematic provision of false information to satisfy external incentives-poses a significant challenge to AI safety. Existing benchmarks often...
RILEC: Detection and Generation of L1 Russian Interference Errors in English Learner Texts
arXiv:2603.07366v1 Announce Type: new Abstract: Many errors in student essays can be explained by influence from the native language (L1). L1 interference refers to errors influenced by a speaker's first language, such as using stadion instead of stadium, reflecting lexical...
Few Tokens, Big Leverage: Preserving Safety Alignment by Constraining Safety Tokens during Fine-tuning
arXiv:2603.07445v1 Announce Type: new Abstract: Large language models (LLMs) often require fine-tuning (FT) to perform well on downstream tasks, but FT can induce safety-alignment drift even when the training dataset contains only benign data. Prior work shows that introducing a...
Cross-Modal Taxonomic Generalization in (Vision-) Language Models
arXiv:2603.07474v1 Announce Type: new Abstract: What is the interplay between semantic representations learned by language models (LM) from surface form alone to those learned from more grounded evidence? We study this question for a scenario where part of the input...
KohakuRAG: A simple RAG framework with hierarchical document indexing
arXiv:2603.07612v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) systems that answer questions from document collections face compounding difficulties when high-precision citations are required: flat chunking strategies sacrifice document structure, single-query formulations miss relevant passages through vocabulary mismatch, and single-pass inference...
Whitening Reveals Cluster Commitment as the Geometric Separator of Hallucination Types
arXiv:2603.07755v1 Announce Type: new Abstract: A geometric hallucination taxonomy distinguishes three failure types -- center-drift (Type~1), wrong-well convergence (Type~2), and coverage gaps (Type~3) -- by their signatures in embedding cluster space. Prior work found Types~1 and~2 indistinguishable in full-dimensional contextual...
Benchmarking Large Language Models for Quebec Insurance: From Closed-Book to Retrieval-Augmented Generation
arXiv:2603.07825v1 Announce Type: new Abstract: The digitization of insurance distribution in the Canadian province of Quebec, accelerated by legislative changes such as Bill 141, has created a significant "advice gap", leaving consumers to interpret complex financial contracts without professional guidance....
vLLM Hook v0: A Plug-in for Programming Model Internals on vLLM
arXiv:2603.06588v1 Announce Type: new Abstract: Modern artificial intelligence (AI) models are deployed on inference engines to optimize runtime efficiency and resource allocation, particularly for transformer-based large language models (LLMs). The vLLM project is a major open-source library to support model...
Switchable Activation Networks
arXiv:2603.06601v1 Announce Type: new Abstract: Deep neural networks, and more recently large-scale generative models such as large language models (LLMs) and large vision-action models (LVAs), achieve remarkable performance across diverse domains, yet their prohibitive computational cost hinders deployment in resource-constrained...
Scale Dependent Data Duplication
arXiv:2603.06603v1 Announce Type: new Abstract: Data duplication during pretraining can degrade generalization and lead to memorization, motivating aggressive deduplication pipelines. However, at web scale, it is unclear what constitutes a ``duplicate'': beyond surface-form matches, semantically equivalent documents (e.g. translations) may...