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
VQIL: Visible Quantum Interference Learning for Perturbation Response Ranking on Graphs Introduces a parameterized quantum-walk model optimized via backpropagation to rank network node responses to seed perturbations. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Proposes a quantum walk method for graph ranking, not an LLM benchmark or evaluation.
Transformer-based In-Context Score Operator Learning for Diffusion Models Proposes an in-context score operator framework allowing diffusion models to adapt to unseen target distributions without retraining. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Focuses on score operator learning for diffusion adaptation, not compression or efficient serving.
Transformer-based In-Context Score Operator Learning for Diffusion Models Proposes an in-context score operator framework allowing diffusion models to adapt to unseen target distributions without retraining. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Presents an in-context operator framework for diffusion models, not LLM benchmarking.
Transformer-based In-Context Score Operator Learning for Diffusion Models Proposes an in-context score operator framework allowing diffusion models to adapt to unseen target distributions without retraining. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Theoretical diffusion operator learning framework unrelated to AI safety or guardrails.
Towards Understanding the Training of GFlowNets: A Theoretical Perspective Develops a theoretical framework linking training objective losses to sampling accuracy and finite-time convergence in Generative Flow Networks. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Mathematical convergence theory for GFlowNets unrelated to deployment safety or alignment.
Towards Understanding the Training of GFlowNets: A Theoretical Perspective Develops a theoretical framework linking training objective losses to sampling accuracy and finite-time convergence in Generative Flow Networks. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Theoretical analysis of GFlowNets convergence, not language model evaluation.
Towards Understanding the Training of GFlowNets: A Theoretical Perspective Develops a theoretical framework linking training objective losses to sampling accuracy and finite-time convergence in Generative Flow Networks. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Addresses theoretical sampling bounds of GFlowNets, not small models or efficient inference.
Machine Unlearning for Federated Socially Fair k-Means Clustering Presents an exact machine unlearning method for federated socially fair k-means clustering using separate group-level local summaries. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Addresses machine unlearning in federated clustering, not language model evaluation.
Machine Unlearning for Federated Socially Fair k-Means Clustering Presents an exact machine unlearning method for federated socially fair k-means clustering using separate group-level local summaries. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Focuses on unlearning speed in federated clustering rather than lightweight LLM inference.
SeoulMMOD: A Large-Scale Multimodal Origin-Destination Flow Benchmark Introduces a multi-mode urban mobility benchmark across Seoul to evaluate origin-destination travel demand forecasting models. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Focuses on urban mobility spatio-temporal modeling, not efficient inference or model compression.
SeoulMMOD: A Large-Scale Multimodal Origin-Destination Flow Benchmark Introduces a multi-mode urban mobility benchmark across Seoul to evaluate origin-destination travel demand forecasting models. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Benchmark is for urban origin-destination traffic flow forecasting, not language model evaluation.
SeoulMMOD: A Large-Scale Multimodal Origin-Destination Flow Benchmark Introduces a multi-mode urban mobility benchmark across Seoul to evaluate origin-destination travel demand forecasting models. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Urban traffic forecasting benchmark unrelated to AI safety, guardrails, or alignment.
SmolWAM: A Latent World-Action Model for Robotic Manipulation on Edge Devices Presents a compressed latent world-action model enabling faster robotic manipulation inference on resource-constrained edge hardware. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Focuses on robotic edge-device latency and compute reduction, not AI safety or guardrails.
SmolWAM: A Latent World-Action Model for Robotic Manipulation on Edge Devices Presents a compressed latent world-action model enabling faster robotic manipulation inference on resource-constrained edge hardware. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Proposes a lightweight robotic policy model rather than an LLM evaluation benchmark.
SmolWAM: A Latent World-Action Model for Robotic Manipulation on Edge Devices Presents a compressed latent world-action model enabling faster robotic manipulation inference on resource-constrained edge hardware. Preprint	Computer vision for industrial applications via arxiv · 12 Oct, 16:29	Not relevant Targets general robotic manipulation on edge devices, not industrial inspection or defect detection.
Studying the Effects of Task Diversity Using Small Mathematical Worlds Introduces two synthetic mathematical task datasets to study how task diversity affects model learning during pretraining. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Investigates task diversity's effect on internal representations, not efficient models or compression.
Studying the Effects of Task Diversity Using Small Mathematical Worlds Introduces two synthetic mathematical task datasets to study how task diversity affects model learning during pretraining. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Introduces synthetic math environments to analyze training dynamics, not LLM benchmark evaluation.
Studying the Effects of Task Diversity Using Small Mathematical Worlds Introduces two synthetic mathematical task datasets to study how task diversity affects model learning during pretraining. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Studies internal representations in synthetic settings, not deployment safety or robustness.
How Well Do Prior-Data Fitted Networks Amortize Bayesian Inference? Empirically evaluates the Bayesian posterior approximation quality and amortization errors of prior-data fitted networks across synthetic prior tasks. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Deals with Bayesian posterior predictive distribution error, not AI safety or deployment guardrails.
How Well Do Prior-Data Fitted Networks Amortize Bayesian Inference? Empirically evaluates the Bayesian posterior approximation quality and amortization errors of prior-data fitted networks across synthetic prior tasks. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Analyzes Bayesian approximation error in Prior-Data Fitted Networks, not LLM benchmarks.
How Well Do Prior-Data Fitted Networks Amortize Bayesian Inference? Empirically evaluates the Bayesian posterior approximation quality and amortization errors of prior-data fitted networks across synthetic prior tasks. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Examines Bayesian inference error decomposition rather than model compression or efficient serving.
From Pixels to Privacy-Sensitive Features: An Information-Theoretic Framework for Reconstruction Bounds Introduces an information-theoretic framework using rate-distortion theory to establish computable lower bounds on feature reconstruction under DP-SGD. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Focuses on differential privacy reconstruction bounds rather than LLM evaluation or benchmarks.
From Pixels to Privacy-Sensitive Features: An Information-Theoretic Framework for Reconstruction Bounds Introduces an information-theoretic framework using rate-distortion theory to establish computable lower bounds on feature reconstruction under DP-SGD. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Investigates privacy distortion bounds under DP-SGD, not model compression or efficient serving.
Geometry-Aware Multi-Objective Optimization: Norm-Constrained Oracles with Entropic Weight Updates Analyzes convergence rates of a multi-objective optimization method pairing norm-constrained linear minimization oracles with entropic task-weight updates. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Examines multi-objective optimization algorithms, not language model evaluation or benchmarks.
Geometry-Aware Multi-Objective Optimization: Norm-Constrained Oracles with Entropic Weight Updates Analyzes convergence rates of a multi-objective optimization method pairing norm-constrained linear minimization oracles with entropic task-weight updates. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Concerns optimization theory rather than small language models, quantization, or inference serving.
Geometry-Aware Multi-Objective Optimization: Norm-Constrained Oracles with Entropic Weight Updates Analyzes convergence rates of a multi-objective optimization method pairing norm-constrained linear minimization oracles with entropic task-weight updates. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Theoretical multi-objective descent paper unrelated to AI safety, hallucination, or guardrails.
Optimal Algorithms for Nonsmooth Nonconvex Stochastic Optimization with Constraints Establishes stochastic first-order and zeroth-order algorithms achieving optimal rates for constrained nonsmooth nonconvex optimization. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Focuses on pure mathematical optimization algorithms, not model compression or efficient serving.
Optimal Algorithms for Nonsmooth Nonconvex Stochastic Optimization with Constraints Establishes stochastic first-order and zeroth-order algorithms achieving optimal rates for constrained nonsmooth nonconvex optimization. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Mathematical optimization theory with no focus on AI safety, robustness, or reliability.
Optimal Algorithms for Nonsmooth Nonconvex Stochastic Optimization with Constraints Establishes stochastic first-order and zeroth-order algorithms achieving optimal rates for constrained nonsmooth nonconvex optimization. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Presents theoretical rates for non-convex stochastic optimization, not LLM evaluation.
Randomized Momentum Transport with Cheap Predictors Introduces randomized momentum transport to track stochastic gradients using probabilistically sampled transport corrections alongside cheap predictors. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Theoretical optimization study without direct application to AI safety, guardrails, or alignment.
Randomized Momentum Transport with Cheap Predictors Introduces randomized momentum transport to track stochastic gradients using probabilistically sampled transport corrections alongside cheap predictors. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Covers theoretical momentum optimization algorithms rather than LLM benchmarking or evaluation.
Randomized Momentum Transport with Cheap Predictors Introduces randomized momentum transport to track stochastic gradients using probabilistically sampled transport corrections alongside cheap predictors. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Examines optimization theory and gradient oracle costs, not small models or inference acceleration.
Adaptive Variance Reduction without Knowing Oracle Smoothness Proposes a two-phase variance reduction method that adapts to average, standard, and noiseless smoothness settings without predefined hyperparameter constants. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant A theoretical optimization paper on variance reduction, not language model evaluation.
Adaptive Variance Reduction without Knowing Oracle Smoothness Proposes a two-phase variance reduction method that adapts to average, standard, and noiseless smoothness settings without predefined hyperparameter constants. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Concerns stochastic non-convex optimization convergence, not model compression or efficient serving.
Adaptive Variance Reduction without Knowing Oracle Smoothness Proposes a two-phase variance reduction method that adapts to average, standard, and noiseless smoothness settings without predefined hyperparameter constants. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Theoretical stochastic optimization study unrelated to AI safety or reliability guardrails.
Constrained Probabilistic Numerics: Gaussian Process Solvers for Linear PDEs under Linear Inequality Constraints Develops a sequential conditioning framework enabling Gaussian process linear partial differential equation solvers to enforce linear inequality constraints. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Presents a Gaussian process solver for PDEs, not an LLM benchmark or evaluation study.
Constrained Probabilistic Numerics: Gaussian Process Solvers for Linear PDEs under Linear Inequality Constraints Develops a sequential conditioning framework enabling Gaussian process linear partial differential equation solvers to enforce linear inequality constraints. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Theoretical PDE solver paper with no connection to deployment AI safety or model alignment.
Constrained Probabilistic Numerics: Gaussian Process Solvers for Linear PDEs under Linear Inequality Constraints Develops a sequential conditioning framework enabling Gaussian process linear partial differential equation solvers to enforce linear inequality constraints. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Focuses on probabilistic numerics for PDEs rather than model compression or efficient LLM inference.
Diversity-Aware Ensemble of Quantum-Classical Models via Combinatorial Fusion Evaluates combinatorial fusion analysis to build diverse ensembles combining classical models and variational quantum circuits for classification tasks. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Proposes a quantum-classical model ensemble technique rather than language model evaluation.
Diversity-Aware Ensemble of Quantum-Classical Models via Combinatorial Fusion Evaluates combinatorial fusion analysis to build diverse ensembles combining classical models and variational quantum circuits for classification tasks. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Addresses combinatorial fusion of quantum circuits, not safety, robustness, or reliability guardrails.
Diversity-Aware Ensemble of Quantum-Classical Models via Combinatorial Fusion Evaluates combinatorial fusion analysis to build diverse ensembles combining classical models and variational quantum circuits for classification tasks. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Focuses on quantum-classical ensembling rather than small language models or inference efficiency.
TabMF: Learning Dataset Representations with Tabular Foundation Models Repurposes pretrained tabular foundation models into supervised dataset encoders using dataset-level readouts and contrastive post-training. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Focuses on tabular foundation model embeddings, not model compression or efficient inference techniques.
TabMF: Learning Dataset Representations with Tabular Foundation Models Repurposes pretrained tabular foundation models into supervised dataset encoders using dataset-level readouts and contrastive post-training. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Does not address AI safety, jailbreaks, hallucination mitigation, or deployment guardrails.
TabMF: Learning Dataset Representations with Tabular Foundation Models Repurposes pretrained tabular foundation models into supervised dataset encoders using dataset-level readouts and contrastive post-training. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Studies dataset representation learning rather than LLM benchmarks or evaluation methods.
Shared Canonical Component Removal Beyond Expert Views Presents Shared Canonical Component Removal to quantify predictive information retained by raw machine learning representations beyond predefined expert features. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Presents an audit framework for expert feature extraction rather than deployed AI safety guardrails.
Shared Canonical Component Removal Beyond Expert Views Presents Shared Canonical Component Removal to quantify predictive information retained by raw machine learning representations beyond predefined expert features. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Does not address model compression, small language models, or efficient serving.
Shared Canonical Component Removal Beyond Expert Views Presents Shared Canonical Component Removal to quantify predictive information retained by raw machine learning representations beyond predefined expert features. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Investigates feature representation auditing in clinical models, not LLM evaluation or benchmarks.
On the Design Space of Continuous Diffusion Language Models Investigates training mechanisms in continuous diffusion language models, proposing design modifications for self-conditioning and embedding geometry. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Focuses on diffusion model architecture and sampling efficiency, not safety, guardrails, or alignment.
On the Design Space of Continuous Diffusion Language Models Investigates training mechanisms in continuous diffusion language models, proposing design modifications for self-conditioning and embedding geometry. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Analyzes continuous diffusion architecture design rather than benchmarking or evaluating language models.
Specialists Learn Faster: Knowledge Organization in Pre-training Shapes Knowledge Update Efficiency Examines how chronological versus shuffled pre-training knowledge sequencing influences the sample efficiency of subsequent model knowledge updates. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Investigates pre-training knowledge update dynamics rather than safety, red-teaming, or deployment reliability.
Specialists Learn Faster: Knowledge Organization in Pre-training Shapes Knowledge Update Efficiency Examines how chronological versus shuffled pre-training knowledge sequencing influences the sample efficiency of subsequent model knowledge updates. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Focuses on pre-training data ordering dynamics, not LLM benchmark design or evaluation frameworks.
Returning to Development States: When Full Sandbox Snapshots Help Coding Agents Evaluates full sandbox snapshot exploration for coding agents, finding benefits primarily when recovery requires restoring runtime dependencies outside Git repositories. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Does not study model compression, distillation, quantization, or small on-device models.
Returning to Development States: When Full Sandbox Snapshots Help Coding Agents Evaluates full sandbox snapshot exploration for coding agents, finding benefits primarily when recovery requires restoring runtime dependencies outside Git repositories. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Deals with task retry efficiency for coding agents rather than deployment safety, alignment, or guardrails.
Returning to Development States: When Full Sandbox Snapshots Help Coding Agents Evaluates full sandbox snapshot exploration for coding agents, finding benefits primarily when recovery requires restoring runtime dependencies outside Git repositories. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant It does not introduce or study general LLM evaluation benchmarks or assessment methodology.
Inference-Time Steering Across Scaffolds and Epitopes for Sample-Efficient De Novo Nanobody Design Applies dynamic budget allocation algorithms at inference time to select scaffolds and epitopes during de novo nanobody generative design. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Covers molecular design inference steering, not LLM evaluation benchmarks.
Inference-Time Steering Across Scaffolds and Epitopes for Sample-Efficient De Novo Nanobody Design Applies dynamic budget allocation algorithms at inference time to select scaffolds and epitopes during de novo nanobody generative design. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Focuses on compute steering for molecular generation, not small or efficient language models.
Inference-Time Steering Across Scaffolds and Epitopes for Sample-Efficient De Novo Nanobody Design Applies dynamic budget allocation algorithms at inference time to select scaffolds and epitopes during de novo nanobody generative design. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Computational biology nanobody generation paper, not AI safety or deployment reliability.
Finite-Sample Guarantees for Fully Discrete Conditional Diffusion Models Derives finite-sample convergence guarantees for fully discrete conditional diffusion models with finitely many denoising steps under smooth density assumptions. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:29	Not relevant Statistical theoretical guarantees, not operational AI safety or deployment reliability.
Finite-Sample Guarantees for Fully Discrete Conditional Diffusion Models Derives finite-sample convergence guarantees for fully discrete conditional diffusion models with finitely many denoising steps under smooth density assumptions. Preprint	Small and efficient models via arxiv · 12 Oct, 16:29	Not relevant Mathematical finite-sample theory for conditional diffusion, not practical small model efficiency.
Finite-Sample Guarantees for Fully Discrete Conditional Diffusion Models Derives finite-sample convergence guarantees for fully discrete conditional diffusion models with finitely many denoising steps under smooth density assumptions. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:29	Not relevant Theoretical statistical estimation bounds for diffusion models, not LLM evaluation.

11,701–11,760 of 13,754← Previous Next →