Every paper checked — nothing hidden

All papers

Every paper checked in the week of 12–18 Oct 2026 · Small and efficient models — nothing is hidden. Brief papers first.

All2,430 In brief3 More (on-topic)455 Not relevant1,957 Not assessed6 Older27 Could not confirm6 Blogs with no paper16 All blogs19

Paper	Topic	Relevance
Beyond Equation Search: Conjecture-Guided Discovery with SCOPE and ClosureBench Introduces a multi-agent framework where simulated scientists formulate, test, and revise scientific conjectures alongside mathematical equations. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on scientific discovery frameworks rather than model compression or efficiency.
Discovering Undisclosed Concepts in LoRA Adapters for Diffusion-Based AIGC Models Proposes an optimization method to discover undisclosed or harmful generation concepts embedded within diffusion model adaptation weights. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Examines concept discovery in LoRAs rather than model compression or inference efficiency.
Evaluating Large Language Models for Multilingual Abstractive Summarization: A Comprehensive Study of English and Bengali Evaluates closed and open models on English and Bengali abstractive summarization using lexical metrics, factuality checks, and automated judges. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Evaluates standard LLMs without proposing efficiency, quantization, or distillation techniques.
Scaling Cheap Synthetic Pretraining Improves Indoor Radio-Map Prediction Improves indoor radio-map prediction by pretraining models on diverse synthetic floorplans and an open-source ray tracer. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies synthetic pretraining for radio maps rather than efficient model compression or serving.
Who Should We Listen to More? Welfare-Aware Preference Acquisition for Pluralistic Alignment Introduces a welfare-aware objective to allocate limited human feedback across diverse individuals when training pluralistic alignment models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, quantization, or efficient inference.
Loss Curvature as a Sampling-Free Proxy for Memorization in Diffusion Models Proposes input loss curvature as a sampling-free metric to detect the onset of training data memorization in diffusion models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on memorization dynamics rather than model compression, distillation, or serving.
Graph Coarsening Enables Escaping Local Optima in Compute-Constrained Nonconvex MAX-CUT Combines spectral graph coarsening with continuous relaxation to find high-quality solutions for compute-constrained MAX-CUT problems. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Deals with combinatorial optimization algorithms rather than small language models or quantization.
Smart Alone, Blind Together: Why Accuracy Fails to CertifyTheory of Mind in Multi-Agent Teams Argues that inferential accuracy fails to evaluate agent coordination safely because correlated errors can compound into costly team actions. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not investigate model pruning, quantization, or efficient deployment.
Revisiting Generative vs. Dense Retrieval: A Static Value Prior Reverses the Gap Shows that incorporating static item value into dense retrieval outperforms generative retrieval in e-commerce recommendation settings. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on retrieval quality comparison rather than model compression or quantization techniques.
Beyond Aggregate Accuracy: Item-Level Transitions in Prompt Optimization Introduces a prompt optimization framework that monitors item-level regressions and recoveries rather than relying solely on aggregate accuracy. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Examines prompt engineering metrics rather than model quantization, distillation, or serving.
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning Presents a large graphical user interface grounding dataset and a combined distillation and reinforcement learning training method for agents. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on grounding accuracy rather than model compression or efficient serving architectures.
Towards a Data-Centric Understanding of Spatial Reasoning in Multimodal Large Language Models Surveys dataset design, construction methods, and evaluation protocols for improving spatial reasoning in multimodal language models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Surveys datasets and evaluation rather than model pruning, quantization, or compression.
From Copying to Composition: How Language Models Develop In-Context Learning Analyzes how language models develop in-context learning by separating simple demonstration copying from multi-token rule composition across training checkpoints. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on context learning dynamics rather than model compression or inference efficiency.
Genesis Steering: Better Activation Steering Through Training Data Steers language model activations using vectors derived directly from the model's own training data or inferred proxy data. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on representation steering rather than model compression, quantization, or efficient serving.
PairIT: Autoregressive Transformers for Low-Data Molecule Optimization The paper introduces a transformer framework trained on matched molecular pairs for prompt-guided, multi-property molecular optimization in low-data regimes. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, efficient serving, or small language models.
Imagining Goals with Geometric Horizon Models for Policy Generalization The paper proposes sampling from a geometric horizon world model to imagine novel outcomes for zero-shot policy generalization. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Not about model compression, quantization, distillation, or efficient inference.
Plant Memory: Technical Language Processing and a Verified Maintenance Ontology for Agent-Accessible Failure Knowledge from SAP PM Work Orders The paper presents a system using technical language processing to transform industrial maintenance work order records into an ontology. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address small language models, quantization, pruning, or serving efficiency.
MA-Bench: Measuring Operational Planning in Language Models The paper introduces a benchmark assessing language models on operational planning from incomplete, complex, and contradictory military document sets. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Benchmark study that does not address model compression, pruning, or efficient inference.
Into the Unknown: Benchmarking AI agents on open scientific research problems The paper introduces a benchmark to evaluate whether frontier language model agents can execute open scientific research and produce follow-up papers. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Assesses frontier research capabilities, not model compression or efficient inference methods.
ListQA: A Benchmark for Evaluating List-Formatted Factual Knowledge Retrieval in Large Language Models The paper introduces an evergreen benchmark to evaluate the ability of language models to retrieve and format multi-element factual lists. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on benchmark creation rather than model compression, quantization, or efficient serving.
Diagnosing Low-Confidence Patient Profiles via Localized Harmful Covariate Shift Detection The paper introduces a diagnostic pipeline to identify low-confidence patient subgroups and detect harmful covariate shifts in clinical prediction models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, distillation, quantization, or efficient deployment.
Generalizing Identifiability Results in Causal Discovery from Complete Data to Missing Data Settings The paper provides theoretical conditions and algorithms for preserving causal graph identifiability when working with missing data distributions. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on causal estimation theory, not model efficiency, distillation, or quantization.
Narrowing the Collaboration Gap, Probably The paper examines communicating calibrated action probabilities rather than direct actions to improve multi-agent collaboration under partial information. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, quantization, distillation, or efficient model serving.
Do Syntax Heads Influence Model Behavior? The paper analyzes whether syntactic attention heads in transformers play a direct causal role in syntactic language predictions. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies attention head interpretability rather than model compression or efficient serving.
Refined Topology-Pass Scheduling: Damped Trust-Region Compensation for Dynamic k-NN Graphs in SE(3)-Equivariant Structure Prediction The paper develops an efficient two-pass scheduling method to update nearest-neighbor graphs in SE(3)-equivariant structure prediction models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Optimizes SE(3) graph prediction pipelines rather than efficient language model inference.
MemeTrace: A Multimodal LLM-Based Agent for Safety-Aware Meme Provenance Analysis The paper presents a multimodal agent designed to trace the origin, spread, and potential safety risks of online memes. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on meme analysis using VLMs rather than model compression or efficiency.
Whose Data, Whose Benchmark? A User-Centric Framework for Auditing Evaluation Resources The paper introduces a framework to audit cultural, geographic, and user-representation biases embedded within popular machine learning benchmarks. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on evaluation benchmark auditing rather than model compression or efficient serving.
Exploring Interpretability in Deep Learning-Based Forest Classification - Separating Similar Forest Classes using Sentinel-2 Time Series The paper evaluates recurrent neural networks and feature attribution methods for tree-species classification using satellite time-series imagery. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Discusses remote sensing RNN classification rather than small language models or model compression.
Dimension-Free Scaling Laws for Size Generalization: An Exponential Separation The paper establishes theoretical bounds and sample complexity laws for models generalizing across varying input sizes in structured data. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Discusses theoretical sample complexity for size generalization rather than practical efficient model inference.
Asymptotic Behavior of Transformers The paper mathematically analyzes long-sequence generation in transformers, establishing temperature thresholds required for internal cache convergence. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on asymptotic convergence properties, not model compression or serving.
Steering Vectors for Competing Ethical Frameworks Are Interchangeable: A Competing-Target Test for Activation Steering The paper introduces a test showing that steering vectors designed for competing ethical frameworks behave interchangeably in open-weight models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on activation steering evaluation, not model efficiency or quantization.
Tail Competition Explains Scaling in Best-of-N Sampling for Verifiable Problems The paper analyzes Best-of-N sampling as an extreme-value competition between tails of correct and incorrect responses to explain reward hacking. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies sampling theory rather than model compression or efficient serving.
Seeing, Maintaining, and Learning: The Evolution of Multimodal Agent Memory The paper provides an analytical framework for how multimodal agents form, represent, and adapt memory over long-horizon tasks. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on memory representations rather than small or quantized models.
Beyond Corrected Actions: Training LLM Agents on Textual Critiques The paper proposes a fine-tuning method that trains language model agents using textual expert critiques alongside corrected actions. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on critique-based fine-tuning rather than model pruning or quantization.
Policy-Verified Self-Training: Learning Agent Compliance from Verifiable Policy Rewards The paper introduces a self-training method that uses runtime policy verification feedback as rewards to teach agents safety compliance. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on policy compliance and preference training, not efficiency or quantization.
Evaluating Compressed Memory: Dependence, Utility, and Content Benefit The paper evaluates whether language models genuinely benefit from compressed memory by comparing them against memory-free baseline models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies memory grounding methodology rather than model compression techniques.
ELARSpeech: A 30K-Hour Dataset Across 675 Underrepresented Languages for Long-Tail Multilingual Speech Technologies The paper introduces a speech dataset spanning 675 underrepresented languages to support speech recognition and text-to-speech technologies. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on linguistic coverage, not model compression or serving efficiency.
KoAGE: Do Large Language Models Know, Learn, and Coin Korean Neologisms? The paper introduces a benchmark to evaluate how well language models recognize, infer, and coin newly emerging Korean terms. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on language acquisition evaluation, not model compression or efficiency.
Vision-Language Models Integrate Conflicting Evidence Like Biased Bayesian Observers The paper proposes a Bayesian cue-combination model to explain how vision-language models integrate conflicting visual and textual evidence. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on evidence integration rather than compression or inference efficiency.
How Do LLMs Understand Legal Philosophy? Diagnosing Legal-Philosophical Reasoning in LLMs The paper introduces a benchmark to evaluate how well language models understand legal philosophy and apply it to legal reasoning. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not cover model compression, distillation, or serving efficiency.
MIPS: Mitigating Mode Collapse in Reinforcement Learning for Large Outcome Spaces The paper proposes marginal inverse probability scaling to reduce mode collapse in reinforcement learning with large outcome spaces. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Addresses RL mode collapse, not model quantization or serving efficiency.
Structure-Preserving Document-Level Data Augmentation via Discourse Parsing The paper introduces a document-level data augmentation method using discourse structure parsing to preserve logical flow during text generation. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Not focused on model compression, quantization, or efficient inference.
Retrieve What Helps, Only When It Helps: Agentic Reinforcement Learning for Selective Retrieval-Augmented Machine Translation The paper proposes a reinforcement learning framework that decides when and what external knowledge to retrieve for machine translation. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Addresses selective retrieval rather than model pruning, distillation, or serving.
Typed Decision Models on Tabular Data The paper compares typed probabilistic decision models with large language models on tabular data classification and format robustness. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on decision formatting robustness rather than model compression techniques.
Evidence-Anchored Overdose Death Prediction from Death Investigation Reports The paper presents an automated framework to predict suspected drug overdose deaths from investigation reports using expert-guided rationales. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model efficiency, distillation, or quantization.
TEX-MEX: Similarity-Guided Temporal Contrastive Learning for Time Series The paper presents a contrastive learning method for time series that unifies morphological similarity and temporal proximity. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Addresses self-supervised time series learning rather than LLM inference optimization.
$τ$-Emotion-Bench: Exposing the Interaction Fragility of LLM Agents through Real-World User Emotions The paper introduces a benchmark to measure how tool-calling agents perform when interacting with emotionally varied, impatient users. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on agent evaluation and training rather than model compression or efficiency.
REWIND: Uncertainty-Guided Trajectory Repair for Test-Time Reasoning The paper introduces a training-free controller that uses token entropy to target and regenerate faulty reasoning steps at test time. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, quantization, or lightweight model design.
Learning Across Temporal Graph Representation Silos The paper studies transferring learned predictive structure between continuous-time dynamic graphs and temporal knowledge graphs across distinct representations. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Investigates graph pretraining transfer rather than LLM efficiency or distillation.
Who's Behind the Text? Robust Model Authorship Attribution The paper proposes a likelihood-based attribution method to identify the source model of machine-generated text under distribution shifts. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address small models, distillation, or inference optimization.
L3M: A Verifiable Harness and Runtime for Self-Extending and Self-Improving Agents The paper presents a Lean 4 runtime allowing self-extending agents to run verified computations across tools without constant human oversight. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Addresses system runtime engineering rather than model compression or inference efficiency.
Beyond Static Positional Encoding: Bayesian Time-aware Attention for Non-Stationary Long-term Time Series Forecasting The paper proposes a Bayesian formulation of self-attention to address non-stationary temporal dependencies in long-term time series forecasting. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on forecasting accuracy rather than model compression, quantization, or serving efficiency.
Sparse Autoencoders Are a Shared Coordinate System for Symbolic Matching The paper uses sparse autoencoders to establish a shared coordinate system for matching symbols across heterogeneous neuro-symbolic systems. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Uses sparse autoencoders for interpretability, not model compression or serving.
Model What Matters: Decision-Guided Selective World Modeling for Language Agents The paper proposes a decision-guided co-training method that selectively filters observation units to improve world modeling for language model agents. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on selective world modeling objectives, not model compression.
HACHIMI-II: From Role-Play to Learner Behavior Modeling for Faithful Student Simulation The paper simulates student behavior by modeling individual learner history and constraining simulated actions via a third-person prediction harness. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on behavioral simulation harnesses, not small models or quantization.
Measuring and Reducing Unusable Replies of LLM Agents The paper measures whether requester users can understand multi-step agent replies and evaluates how explicit constraints improve reply usability. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Addresses communication quality in agent outputs, not model efficiency or distillation.
MCAIR: Monitoring and Correcting Agent Unfaithfulness via Internal Representations The paper proposes a runtime monitoring method to detect and correct language model agents that deviate from intended procedures without reporting it. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on internal representation probing, not model compression or efficiency.
Interlock: From Local Correctness to Task Completion in Long-Horizon Agents The paper proposes a graph-based task synthesis framework to help long-horizon agents satisfy inter-dependent constraints and achieve complete task outcomes. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on task synthesis and evaluation, not efficient inference or pruning.
Learning Individual Scientific Taste with Privileged Hindsight Distillation The paper models individual scientific taste using criterion assessments and introduces privileged hindsight distillation to align proposals with researcher preferences. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on preference distillation, not small efficient models.
Articulation Upper Bounds for Language Models The paper introduces prompt optimization and recovery measurement tools to evaluate the limits of transmitting concepts between scaling language models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Examines scaling behavior, not model compression or efficiency.

1,021–1,080 of 2,430← Previous Next →