Every paper checked — nothing hidden

All papers

Every paper checked for brief #4 of 12–19 Oct 2026 — nothing is hidden. Brief papers first.

All17,313 In brief24 More (on-topic)3,333 Not relevant13,754 Not assessed87 Older352 Could not confirm93 Blogs with no paper266 All blogs284

Paper	Topic	Relevance
When Sparsity Breaks Connectivity: Functional Reachability and Guided Topology Dynamics in Extreme Sparse Training The authors develop a reachability-guided dynamic sparse training method that preserves functional pathways during extreme network pruning and regrowth. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Studies sparse training dynamics rather than LLM evaluation or benchmark design.
When Sparsity Breaks Connectivity: Functional Reachability and Guided Topology Dynamics in Extreme Sparse Training The authors develop a reachability-guided dynamic sparse training method that preserves functional pathways during extreme network pruning and regrowth. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on extreme pruning and sparsity rather than deployment safety, hallucinations, or jailbreaks.
Monarch-Muon: Structure-Aware Muon for Fast and Memory-Efficient Pretraining The authors propose an optimizer for Monarch-factored weights that replaces dense matrix orthogonalization with smaller block updates to reduce memory and pretraining time. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on optimization efficiency during pretraining rather than safety, alignment, or guardrails.
Monarch-Muon: Structure-Aware Muon for Fast and Memory-Efficient Pretraining The authors propose an optimizer for Monarch-factored weights that replaces dense matrix orthogonalization with smaller block updates to reduce memory and pretraining time. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Presents an optimization algorithm rather than an LLM evaluation framework or benchmark.
Formal-Verification-Guided Bayesian Optimization for Non-Intrusive Last-Mile Coverage Closure The authors present a verification workflow combining reachability analysis with Bayesian optimization to close residual test bins in chip design verification. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Deals with semiconductor functional verification, not AI safety guardrails or reliability in deployment.
Formal-Verification-Guided Bayesian Optimization for Non-Intrusive Last-Mile Coverage Closure The authors present a verification workflow combining reachability analysis with Bayesian optimization to close residual test bins in chip design verification. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Discusses Bayesian optimization for chip verification, not small language models or efficient inference.
Formal-Verification-Guided Bayesian Optimization for Non-Intrusive Last-Mile Coverage Closure The authors present a verification workflow combining reachability analysis with Bayesian optimization to close residual test bins in chip design verification. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Focuses on hardware verification coverage closure rather than language model evaluation.
On Sleep-Wake Disorder Identification from Polysomnographic Representations The authors compare self-supervised and sleep-stage-supervised representations for identifying sleep-wake disorders across thousands of polysomnography recordings. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on clinical diagnostic benchmarking rather than AI safety guardrails or LLM reliability.
On Sleep-Wake Disorder Identification from Polysomnographic Representations The authors compare self-supervised and sleep-stage-supervised representations for identifying sleep-wake disorders across thousands of polysomnography recordings. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Evaluates sleep polysomnography models rather than LLM benchmarks or evaluation methods.
On Sleep-Wake Disorder Identification from Polysomnographic Representations The authors compare self-supervised and sleep-stage-supervised representations for identifying sleep-wake disorders across thousands of polysomnography recordings. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Examines medical representation learning, not model compression or efficient serving.
Effect of Spherical Parameter Space on Number of Non-Forgetting Tasks and Capacity of Continual Learning The authors analyze continual learning on a spherical parameter space to determine how many tasks a model can learn without catastrophic forgetting. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Theoretical paper on task non-forgetting without focus on deployment safety or alignment.
Effect of Spherical Parameter Space on Number of Non-Forgetting Tasks and Capacity of Continual Learning The authors analyze continual learning on a spherical parameter space to determine how many tasks a model can learn without catastrophic forgetting. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Analyzes spherical parameter geometry in continual learning, not small models or efficient inference.
Effect of Spherical Parameter Space on Number of Non-Forgetting Tasks and Capacity of Continual Learning The authors analyze continual learning on a spherical parameter space to determine how many tasks a model can learn without catastrophic forgetting. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Theoretical analysis of continual learning parameter spaces, not LLM evaluation methods.
Selective Inference for Clustering Under Unknown Covariance The authors introduce a selective hypothesis test for comparing cluster means under an unknown covariance matrix without inflating false-positive rates. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Covers post-clustering statistical inference, not language model evaluation or benchmarks.
Selective Inference for Clustering Under Unknown Covariance The authors introduce a selective hypothesis test for comparing cluster means under an unknown covariance matrix without inflating false-positive rates. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on classical statistical hypothesis testing rather than AI system safety and reliability.
Selective Inference for Clustering Under Unknown Covariance The authors introduce a selective hypothesis test for comparing cluster means under an unknown covariance matrix without inflating false-positive rates. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Concerns theoretical statistics for clustering, not efficient model inference or compression.
From Orbit Matching to Docking: A Hierarchical Reinforcement-Learning Approach to Multiphase Rendezvous Trajectory Design The paper proposes a hierarchical reinforcement learning framework to automate multi-phase spacecraft rendezvous and docking trajectories. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Does not benchmark or evaluate language models.
From Orbit Matching to Docking: A Hierarchical Reinforcement-Learning Approach to Multiphase Rendezvous Trajectory Design The paper proposes a hierarchical reinforcement learning framework to automate multi-phase spacecraft rendezvous and docking trajectories. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Deals with spacecraft control dynamics rather than AI deployment safety or language model alignment.
From Orbit Matching to Docking: A Hierarchical Reinforcement-Learning Approach to Multiphase Rendezvous Trajectory Design The paper proposes a hierarchical reinforcement learning framework to automate multi-phase spacecraft rendezvous and docking trajectories. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on reinforcement learning for spacecraft docking, not small or efficient language models.
From Orbit Matching to Docking: A Hierarchical Reinforcement-Learning Approach to Multiphase Rendezvous Trajectory Design The paper proposes a hierarchical reinforcement learning framework to automate multi-phase spacecraft rendezvous and docking trajectories. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on spacecraft guidance trajectories, not industrial visual inspection or defect detection.
Structure-preserving fidelity diagnostics for partial differential equations The paper introduces diagnostic scores to test whether neural network surrogates respect underlying physical conservation laws rather than numerical errors. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Evaluates physical conservation properties in PDE surrogates, not AI deployment safety.
Structure-preserving fidelity diagnostics for partial differential equations The paper introduces diagnostic scores to test whether neural network surrogates respect underlying physical conservation laws rather than numerical errors. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on physics-informed PDE surrogates rather than small model compression or serving.
Structure-preserving fidelity diagnostics for partial differential equations The paper introduces diagnostic scores to test whether neural network surrogates respect underlying physical conservation laws rather than numerical errors. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Develops physics diagnostic metrics for differential equations rather than LLM benchmarks.
A Time-Synchronous Prequential Evaluation Framework for Next-Week Student Disengagement Prediction from Moodle LMS Logs The paper evaluates online machine learning models within a time-synchronous framework to predict weekly student disengagement from activity logs. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on educational log classification rather than model compression or efficient serving.
A Time-Synchronous Prequential Evaluation Framework for Next-Week Student Disengagement Prediction from Moodle LMS Logs The paper evaluates online machine learning models within a time-synchronous framework to predict weekly student disengagement from activity logs. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Studies student disengagement prediction rather than language model evaluation or benchmarks.
A Time-Synchronous Prequential Evaluation Framework for Next-Week Student Disengagement Prediction from Moodle LMS Logs The paper evaluates online machine learning models within a time-synchronous framework to predict weekly student disengagement from activity logs. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Addresses student disengagement models rather than AI deployment safety or guardrails.
Longitudinal Learner Participation Profiling in Online Learning: Heterogeneous Consensus Clustering under Unequal Observation Windows The paper evaluates clustering methods on learning management system logs to identify learner participation profiles across unequal observation windows. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Covers learner analytics rather than AI deployment safety, guardrails, or hallucination.
Longitudinal Learner Participation Profiling in Online Learning: Heterogeneous Consensus Clustering under Unequal Observation Windows The paper evaluates clustering methods on learning management system logs to identify learner participation profiles across unequal observation windows. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Studies student log clustering rather than small, efficient model compression or serving.
Longitudinal Learner Participation Profiling in Online Learning: Heterogeneous Consensus Clustering under Unequal Observation Windows The paper evaluates clustering methods on learning management system logs to identify learner participation profiles across unequal observation windows. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Analyzes student participation clustering rather than language model evaluation or benchmarking.
Merit: Making every modality earn its contribution in multimodal attention decoding The paper introduces a multimodal auditory attention decoding framework designed to maintain reliance on neural signals alongside behavioural tracking. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on auditory signal decoding rather than AI deployment safety or guardrails.
Merit: Making every modality earn its contribution in multimodal attention decoding The paper introduces a multimodal auditory attention decoding framework designed to maintain reliance on neural signals alongside behavioural tracking. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Addresses hearing aid decoding methods rather than small model compression or inference.
Merit: Making every modality earn its contribution in multimodal attention decoding The paper introduces a multimodal auditory attention decoding framework designed to maintain reliance on neural signals alongside behavioural tracking. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Evaluates auditory attention decoding inputs rather than language model benchmarking methods.
ATTUNE: Agentic Throughput Tuning for LLM Training The paper presents an automated coding agent that tunes large model training configurations for throughput under user-defined compute budgets. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Addresses training throughput optimization rather than AI deployment safety or guardrails.
ATTUNE: Agentic Throughput Tuning for LLM Training The paper presents an automated coding agent that tunes large model training configurations for throughput under user-defined compute budgets. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Focuses on tuning distributed training throughput rather than LLM evaluation benchmarks.
Risk-Averse Inference-Time Diffusion Model Alignment with Finite-Sample Guarantees The paper proposes an inference-time alignment method that optimizes expected costs while controlling tail risk in diffusion models. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Analyzes diffusion alignment algorithms rather than language model evaluation benchmarks.
Risk-Averse Inference-Time Diffusion Model Alignment with Finite-Sample Guarantees The paper proposes an inference-time alignment method that optimizes expected costs while controlling tail risk in diffusion models. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on diffusion inference alignment rather than model compression or efficient serving.
Training Generalist Navigation Policies via Action Self-Distillation The paper introduces a self-distillation framework where a vision-language model teaches itself language-conditioned robotic navigation policies without human annotations. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on general mobile robot navigation rather than industrial manufacturing inspection.
Training Generalist Navigation Policies via Action Self-Distillation The paper introduces a self-distillation framework where a vision-language model teaches itself language-conditioned robotic navigation policies without human annotations. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Presents a robot navigation policy rather than an LLM evaluation benchmark.
Training Generalist Navigation Policies via Action Self-Distillation The paper introduces a self-distillation framework where a vision-language model teaches itself language-conditioned robotic navigation policies without human annotations. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Uses distillation for robot policy learning rather than model compression or efficient serving.
Training Generalist Navigation Policies via Action Self-Distillation The paper introduces a self-distillation framework where a vision-language model teaches itself language-conditioned robotic navigation policies without human annotations. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on robotic navigation learning rather than AI deployment safety or guardrails.
FedLoRD: Personalized Federated Learning based on Low-Rank Structures of Model Parameter Differences The paper introduces a personalized federated learning method that regularizes local model updates using low-rank difference structures. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Studies federated optimization rather than AI deployment safety, guardrails, or hallucination.
FedLoRD: Personalized Federated Learning based on Low-Rank Structures of Model Parameter Differences The paper introduces a personalized federated learning method that regularizes local model updates using low-rank difference structures. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Focuses on personalized federated learning rather than language model evaluation frameworks.
FedLoRD: Personalized Federated Learning based on Low-Rank Structures of Model Parameter Differences The paper introduces a personalized federated learning method that regularizes local model updates using low-rank difference structures. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Targets federated regularizers rather than small model compression, distillation, or serving.
Mixed Precision Scaling Laws for Quantized Pretraining The paper proposes mixed precision scaling laws to model the trade-offs of low-precision quantization during model pretraining. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Analyzes quantization scaling behavior rather than AI deployment safety or guardrails.
Mixed Precision Scaling Laws for Quantized Pretraining The paper proposes mixed precision scaling laws to model the trade-offs of low-precision quantization during model pretraining. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Studies scaling behavior under quantization rather than general LLM evaluation benchmarks.
textGrain watermarking is harder to detect: can sum-tests still do it optimally? The paper analyzes the theoretical detection limits and sum-test efficiency of textGrain watermarking under human text edits. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on watermarking theory rather than model compression or efficient inference methods.
textGrain watermarking is harder to detect: can sum-tests still do it optimally? The paper analyzes the theoretical detection limits and sum-test efficiency of textGrain watermarking under human text edits. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Analyzes statistical watermark detection boundaries rather than general LLM evaluation benchmarks.
Focus on What You Need: Global-to-Local Dynamic Attention for Multimodal Sarcasm Detection The paper introduces a global-to-local dynamic attention model to filter visual noise and improve multimodal sarcasm detection. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Presents a multimodal sarcasm classifier rather than an LLM evaluation benchmark.
Focus on What You Need: Global-to-Local Dynamic Attention for Multimodal Sarcasm Detection The paper introduces a global-to-local dynamic attention model to filter visual noise and improve multimodal sarcasm detection. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on multimodal attention mechanisms rather than model compression or efficient serving.
Focus on What You Need: Global-to-Local Dynamic Attention for Multimodal Sarcasm Detection The paper introduces a global-to-local dynamic attention model to filter visual noise and improve multimodal sarcasm detection. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Studies sarcasm detection rather than AI deployment safety, hallucination, or guardrails.
Confine to the Task, Anchor the Rest: Concept-Guided Retention Regularizer for Parameter-Efficient Fine-Tuning The paper proposes a concept-guided regularizer using sparse autoencoders to reduce catastrophic forgetting during parameter-efficient fine-tuning. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Proposes a regularizer for parameter-efficient tuning rather than an LLM evaluation benchmark.
Confine to the Task, Anchor the Rest: Concept-Guided Retention Regularizer for Parameter-Efficient Fine-Tuning The paper proposes a concept-guided regularizer using sparse autoencoders to reduce catastrophic forgetting during parameter-efficient fine-tuning. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on retaining knowledge during fine-tuning rather than model compression or inference efficiency.
MissPath-FM: Flow Matching with Structured Priors for Partially Observed Time Series The paper develops a flow matching framework with structured priors to model partially observed time series without wasting capacity on observed entries. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Addresses time series flow matching rather than small, efficient model serving or compression.
MissPath-FM: Flow Matching with Structured Priors for Partially Observed Time Series The paper develops a flow matching framework with structured priors to model partially observed time series without wasting capacity on observed entries. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Evaluates time series imputation rather than language model evaluation or benchmarking.
MissPath-FM: Flow Matching with Structured Priors for Partially Observed Time Series The paper develops a flow matching framework with structured priors to model partially observed time series without wasting capacity on observed entries. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on missing data imputation rather than deployment safety, guardrails, or hallucination.
Representation Geometry and Inductive Biases in Quantum Classifiers across Classical Visual Benchmarks The paper investigates the representation geometry and inductive biases of quantum classifiers when applied to classical visual benchmark datasets. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Compares quantum classifier inductive biases rather than language model evaluation benchmarks.
Representation Geometry and Inductive Biases in Quantum Classifiers across Classical Visual Benchmarks The paper investigates the representation geometry and inductive biases of quantum classifiers when applied to classical visual benchmark datasets. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Investigates quantum machine learning classification rather than efficient language model compression or inference.
Representation Geometry and Inductive Biases in Quantum Classifiers across Classical Visual Benchmarks The paper investigates the representation geometry and inductive biases of quantum classifiers when applied to classical visual benchmark datasets. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Studies quantum feature mappings rather than AI deployment safety or reliability risks.
Hybrid Parallel Bessel KAN Model: a Kolmogorov Arnold Network: integrated framework for developing a novel architecture to reconstruct the fused-basis approach via a parameter free quadratic interaction The paper proposes a Kolmogorov-Arnold Network architecture using parallel Bessel branches and quadratic interactions to reduce parameters and conditioning problems. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on function approximation architecture rather than AI deployment safety or reliability.
Hybrid Parallel Bessel KAN Model: a Kolmogorov Arnold Network: integrated framework for developing a novel architecture to reconstruct the fused-basis approach via a parameter free quadratic interaction The paper proposes a Kolmogorov-Arnold Network architecture using parallel Bessel branches and quadratic interactions to reduce parameters and conditioning problems. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Evaluates Kolmogorov-Arnold Network variants rather than language model benchmarking methods.

9,301–9,360 of 17,313← Previous Next →