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
PersEvalBench: Can LLM Evaluators Judge Contextually Appropriate Use of Persona Information? The paper introduces a benchmark to evaluate whether language models can accurately detect appropriate and inappropriate uses of persona information in dialogue. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address small models, distillation, or quantization.
Truthful and Expected-Payment-Optimal Model Routing in Graph-Based Agentic Workflows The paper formulates task routing across language model providers as an auction mechanism to encourage truthful cost reporting while minimizing context switching. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies routing economics rather than model compression or efficient serving.
TopicMerge: A Divide and Conquer Approach to Topic Modeling The paper introduces a multi-stage topic modeling pipeline that merges fine-grained document clusters using a language model to balance efficiency and quality. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model compression, quantization, or distillation.
Counterfactual Gate-Support Decomposition: Separating Selection from Redistribution in Training-Data Filtering The paper proposes an experimental design that separates example selection from mass redistribution to analyze the effects of training-data quality filtering. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on training data selection methodology rather than model efficiency.
Target-Token Asymmetry Can Masquerade As Depth Specialization The paper shows that differences in predicted target tokens, such as whitespace, can falsely look like layer specialization across different corpora. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on interpretability analysis rather than model compression techniques.
Where to Start, How to Move: Analyzing Curricula under GRPO for Long-Horizon LLM Agents The paper analyzes prefix replay curricula and handoff rules under reinforcement learning to improve long-horizon task completion for language model agents. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, quantization, or efficient inference.
VLX-Go: Robust Embodied Tracking through Multi-Stage Learning and On-Policy Refinement The paper presents a vision-language policy combining pre-aligned representations and multi-stage learning to improve robotic person tracking among distractors. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model compression or inference acceleration.
Learning What to Keep: Transfer-Aware Personal Skill Learning from User Feedback The paper introduces a framework that decomposes user feedback into atomic rules to retain only reusable preferences for personalizing language models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not investigate pruning, distillation, or lightweight models.
The Sum of All Summaries: PROPALIGN for Graded Proposition Salience Evaluation The paper proposes a multiple-summary metric and dataset to evaluate graded proposition salience across diverse text genres. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model compression or efficient serving.
ForkGRPO: Efficient Exploration and Fine-Grained Credit Assignment for Diffusion Language Models The paper introduces a reinforcement learning framework using intermediate-state branching and segment-level credit assignment for diffusion language models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies exploration in diffusion language models, not model efficiency.
More Than Seeing: Vision as Scalable Anchors in Native Multimodal Pre-Training The paper studies how visual input signals affect scaling behavior and non-visual reasoning during native multimodal pretraining. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Examines scaling behavior rather than small efficient model architectures.
ActICL: Post-Training for Agentic Multimodal In-Context Learning The paper proposes an agentic framework enabling multimodal models to identify information gaps and retrieve in-context demonstrations iteratively. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model compression, distillation, or quantization.
What Does LALM Post-training Learn? Rethinking Audio Understanding and Reasoning Capabilities The paper analyzes post-training in audio-language models to separate genuine audio reasoning gains from question-format memorization. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies post-training evaluation artifacts, not model compression or efficiency.
Towards Efficient Simultaneous Inverse Text Normalization with Pretrained Text-to-Text Language Model and Read-Tag-Write Policy The paper develops a streaming end-to-end inverse text normalization architecture using a pretrained text-to-text model with a read-tag-write policy. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Application-specific speech pipeline optimization, not general model efficiency or compression.
Diagnosing and Correcting Reward Hacking in Verifiable Instruction Following The paper analyzes reward hacking in verifiable instruction following and introduces a likelihood bonus method to maintain answer quality. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Addresses reward hacking mitigation, not model compression or efficiency.
A Unified Response Framework for Questions and Tutoring Dialogues The paper presents an item response theory framework to jointly evaluate student performance on standard questions and tutoring dialogue subtasks. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Presents dialogue modeling for education, not efficient model architectures.
GraphRAG-LTM: A Layered Threat Modeling Framework for Attacks on Graph Retrieval-Augmented Generation The paper introduces a threat modeling framework examining adversarial query demotion and graph edit attacks on graph-based RAG systems. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Examines GraphRAG security vulnerabilities, not model compression or efficiency.
Moral Combat: Exploiting Moral Foundations Theory to Jailbreak LLMs The paper introduces a multi-agent framework that reframes harmful requests using Moral Foundations Theory to test model jailbreak vulnerabilities. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on safety vulnerabilities, not model compression or efficiency.
Twenty Days Later: A Pre-Registered Test of History-Based Forecasts of Brand Mentions in Gemini Answers The paper reports a pre-registered test measuring the stability of history-based forecasts for model brand mentions over twenty days. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Investigates output stability rather than model efficiency or distillation.
Are We Really Measuring Forgetting? Pitfalls in Evaluating Machine Unlearning for Multimodal LLMs The paper identifies evaluation pitfalls in multimodal machine unlearning and introduces a benchmark grounded in real-world knowledge. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on evaluating unlearning protocols rather than model compression or acceleration.
J-tau: A Fully Localized Japanese Benchmark for Tool-Agent-User Interaction The paper introduces a localized Japanese benchmark for evaluating tool-using conversational agents in customer-support scenarios. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Introduces an evaluation benchmark rather than efficient model architectures or inference.
Beyond Distribution Matching: Selective Margin Preservation for Supervised Fine-Tuning The paper introduces a soft margin preservation target for supervised fine-tuning to retain non-target token rankings. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Presents a fine-tuning loss rather than pruning, distillation, or quantization.
Share the Symbol, Not the Vector: Training-Free Fact Communication Across Language Models The paper proposes a training-free interface for heterogeneous language models to communicate facts using shared symbolic word indices. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Investigates multi-model hidden state sharing, not compression or inference optimization.
BANGLAPLAN: A Benchmark for Constraint-Based Planning and Decision-Making in Bangla The paper introduces a 1,000-problem benchmark for constraint-based planning and decision-making in the Bangla language. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Presents an evaluation benchmark rather than model compression or efficient serving.
MakSafe: A Spatiotemporal Semantic Framework for Formal Specification and Compliance Checking of Road Vehicle Safety Rules The paper presents a formal spatiotemporal framework to specify and verify road vehicle safety rules across continuous trajectories. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Not about small or efficient language models.
Beyond Reachability: A Relation-Semantic Reasoning Framework for Biomedical Multi-hop QA The paper proposes a relation-semantic reasoning framework to resolve disconnects between answer reachability and valid paths in biomedical graphs. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression or inference efficiency.
Likelihood Is Not Authorship: Can We Reliably Measure Human Contribution in AI-Assisted Writing? The paper reveals vulnerabilities in prompt-predictability metrics used to estimate human contribution in AI-assisted writing. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model efficiency or compression.
Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems The paper investigates how team composition and organizational interdependence affect performance in self-evolving multi-agent systems. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression or serving efficiency.
Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations The paper studies the durability of inference-time value steering in multi-turn conversations when user and model values conflict. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on efficient serving or small models.
MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation The paper presents a diagnostic evaluation framework to measure how inference-time techniques affect parametric knowledge extraction in models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies knowledge extraction techniques, not model compression or efficiency.
EGCForest: Structured Agent Memory Organization for Enterprise Group Chats The paper proposes a multi-view hierarchical memory framework to improve information retrieval from enterprise group chats. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model compression or inference efficiency.
OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score The paper introduces a two-part benchmark to evaluate language models on general and examination knowledge concerning Taiwan. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model efficiency or compression.
SEA-HELM Audio: Evaluating Multilingual ASR-LLM Pipeline Performance Across SEA-HELM Subsets The paper presents an audio evaluation benchmark assessing accuracy drops between speech-input and text-input pipelines in Southeast Asian languages. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression or efficient serving.
RCCM: Diagnosing Gains and Harms in Reasoning-Prompt Transfer The paper analyzes question-level variations and prompt-execution mismatches when transferring reasoning prompts across different language models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Examines prompt transfer and token caps, not model compression.
Beyond Benchmark Accuracy: A Diagnostic Test for LLM Chemical Prediction The paper proposes a diagnostic test hiding target identity to evaluate structure-property chemical reasoning in language models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on efficiency or model compression.
Not All Criteria Matter: Importance-Aware Criteria Weighting for Rubric-Based LLM Evaluation The paper introduces a method that weights rubric criteria through pairwise model comparisons to improve evaluation accuracy. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model efficiency or compression.
MolHop: Scaffold-Hopping Molecular Design with Agentic Reinforcement Learning The paper presents an agentic reinforcement learning framework that navigates organized chemical spaces for cross-scaffold molecular discovery. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model compression or efficient serving.
SoK: Grading Open-Text Legal Answers, from Law School Exams to Trustworthy Rubric Judges The paper systematizes grading requirements for open-text legal exams and compares eight evaluation paradigms using human and model judges. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not cover model compression, quantization, or efficient inference.
Gray-Box Hallucination Detection in LLMs with Token Identities The paper introduces a lightweight gray-box hallucination detector combining token probabilities with vector representations of token identities. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not study model compression, quantization, or efficient serving.
LintEval: Benchmarking LLMs on Linter-Guided File Refactoring The paper presents a multilingual benchmark to evaluate how language models resolve linter alerts while preserving program functionality. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model compression, quantization, or serving.
FACES: Fine-grained Gated Semantic Alignment for Multimodal Emotional Support Conversation The paper introduces a gated semantic alignment framework with multi-source rewards to improve multimodal emotional support conversations. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not cover model pruning, quantization, or efficient serving.
What Is a Clarification? Estimating Model-Dependent Input Ambiguity The paper formalizes clarification criteria to estimate model-dependent input ambiguity based on accessible and valid answers. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, quantization, or efficient serving.
Sparse Autoencoders for Mechanistic Interpretability in NLP: A Critical Survey of Recoverability, Interpretability, and Utility This paper surveys 138 studies on sparse autoencoders in NLP, examining dictionary design, feature recoverability, and practical utility. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Covers interpretability analysis rather than serving or inference efficiency.
Gap Detection: Mitigating Homogenization through Demand and Defensibility This paper introduces Gap Detection, an optimization method that identifies unoccupied semantic directions to reduce homogenization in open-ended model answers. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not cover model compression, pruning, or quantization.
When Occupation-derived Skills Don't Help: On the Limits of Skill-Augmented Career Path Prediction This paper analyzes career path prediction systems, reporting that explicit skill representations provide marginal improvement over text-only job descriptions. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model efficiency, quantization, or distillation.
Repairing Speech Access After Text Updates with Frozen Affine Fields This paper introduces AFFIRM, using affine fields on frozen speech models to repair spoken access to recently updated text knowledge. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not study model compression, pruning, or quantization.
Anuvani-LID: Robust and Calibrated Spoken Language Identification for Indian Languages This paper introduces Anuvani LID, a dedicated 115M parameter model for spoken language identification across 22 Indian languages. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on dedicated conformer LID, not model compression methods.
Neural POS Tagging for the Low-Resource Saraiki This paper introduces an annotated part-of-speech tagging dataset and evaluates pretrained transformer encoders for the low-resource Saraiki language. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Evaluates existing transformer encoders without introducing compression methods.
Taking the PULSE of Medical Reasoning: Step-Level Retrieval Timing for Knowledge-Grounded Inference This paper proposes PUlSE, using internal layer representations to detect factual errors and trigger step-level retrieval during medical reasoning. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on retrieval timing rather than parameter reduction or quantization.
MMLA-RSI: Recursive Self-Improvement through RTT and RTMA This paper proposes MMLA-RSI, a recursive self-improvement framework using reasoning-time policy training and authoritative memory adaptation. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on parameter efficiency or model compression.
Anuvani-Diarization: Zero Shot Speaker Diarization and Annotation Conventions Across 22 Languages This paper presents Anuvani-Diarization, evaluating zero-shot speaker diarization across 22 Indian languages with uncertainty-aware silver supervision. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not explore model distillation, quantization, or efficient architectures.
Do AI Agents Gossip? Evaluative Talk in Human and AI-Agent Online Communities This paper investigates whether gossip discourse naturally emerges in autonomous AI agent online communities compared to human discussion boards. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, quantization, or efficient serving.
Lexical Persistence Without Semantic Entrenchment in AI Agent Communities This paper measures semantic drift and stability across AI agent communities, finding higher drift and lower temporal continuity than human forums. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on linguistic diffusion and community dynamics, not model compression.
From Triage to Impact: A Multi-Hop Reasoning Benchmark for Explainable Persian Hate Speech This paper introduces a multi-hop reasoning benchmark to evaluate fine-grained, explainable hate speech analysis in Persian. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on benchmarking hate speech detection, not model compression or inference efficiency.
Same Claim, Different Interpretations: How LLM Agents Resolve Ambiguity in Scientific Claims This paper proposes Claim Cards to make agent interpretations explicit when verifying ambiguous natural-language scientific claims. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Investigates interpretation variance in LLM agents, not model compression or efficiency.
MolLean: A Process-Verifiable Language for Chemical Reasoning This paper introduces MolLean, a verifiable language and tool interface that grounds chemical reasoning in formal state transitions and explicit checks. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on formal chemical reasoning frameworks rather than model quantization or distillation.
ChronoCulture: Evaluating Cultural Awareness Across Time in Large Language Models This paper introduces ChronoCulture to evaluate how well language models capture temporal changes in cultural attitudes across 72 countries. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Presents an evaluation benchmark rather than small models or inference optimization.
AdaRSI: On-Demand Recursive Self-Improvement for Embodied Agents This paper introduces AdaRSI, an on-demand recursive self-improvement framework for embodied agents that isolates and fine-tunes only persistent sub-step failures. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on targeted sub-step policy fine-tuning rather than model compression or inference efficiency.
Lost in Translation: Why Lower Bias Scores Don't Mean Fairer Models This paper evaluates multilingual bias audits across ten language models, showing that lower stereotype scores often reflect translation artifacts rather than fairer behavior. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Investigates multilingual bias metrics, not model compression or efficient serving.
The Agentic Scientist: Orchestrating Coding Agents for Automated Scientific (Co-)Discovery Presents an end-to-end distributed system orchestrating coding agents to execute computational research experiments alongside human scientists. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on agent orchestration rather than efficient model architectures or compression.

961–1,020 of 2,430← Previous Next →