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
Lüshi-1840: Building and Probing a Knowledge Cutoff for a Classical Chinese Language Model This paper introduces Lüshi-1840, a pre-modern Chinese language model trained on historical texts to evaluate knowledge cutoffs and hindsight bias. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on temporal dataset curation rather than model compression.
Deep Speech Representations Track Word Timing along a Low-Dimensional Circular Trajectory This paper demonstrates that deep speech representations track word timing along low-dimensional circular trajectories across multiple model architectures. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Studies acoustic representation geometry, not model compression or serving.
RelLedger: Relational Information Governance for LLM Agent Organizations This paper introduces RelLedger, an auditing framework to trace how relational histories drive reporting choices and actions in multi-agent systems. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Addresses agent audit mechanisms rather than model compression or inference efficiency.
ATTCK-R: Knowledge-Aware Long-Horizon Jailbreaking for Protected LLMs This paper presents ATTCK-R, a planning-driven multi-turn red teaming framework to evaluate language models guarded by prompt safety defenses. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Addresses safety red-teaming rather than model efficiency or compression.
Strong yet Controllable LLM Abstention without Learning to Abstain This paper proposes TAPS, combining standard accuracy training with post-hoc confidence selection to give controllable abstention rates without specialized tuning. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Investigates selective answering methods rather than compression or efficient serving.
Rule vs.Consequence: Dissociable Internal Representations of Moral Reasoning in LLMs This paper uses activation steering and probing to identify distinct internal representations for rule-based versus consequence-based moral reasoning in language models. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Mechanistic interpretability study, not model compression or efficient serving research.
PersonaFactor: Preference Factors with Dynamic Selection for LLM Personalization This paper presents PersonaFactors, a framework that distills reusable preference patterns across tasks for dynamic prompt personalization in language models. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Does not address model compression, quantization, distillation, or efficient serving.
CultureBot: A Framework for Retrieval-Augmented Exploration of Cultural Heritage Data This paper introduces an interactive retrieval-augmented framework that combines knowledge graphs and text retrieval to explore digital cultural heritage archives. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Develops a domain retrieval application without focusing on model compression techniques.
CivWorldBench: Evaluating Language World Models in Structured Multi-Agent Sandbox Worlds This paper introduces a multi-agent sandbox benchmark to evaluate language models on state inference, forecasting, and intervention prediction. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Presents a forecasting benchmark rather than model compression or efficient serving.
Differential Privacy Protected Preference Alignment on Survey Derived Data This paper presents a label differential privacy framework to protect survey-derived ordinal preference data during language model alignment. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Studies privacy guarantees during training, not inference speedup or model compression.
The interpreter as developer: An autoethnographic design study of building a computer-assisted interpreting tool through human-LLM co-creation This paper presents a case study of an interpreter co-creating a local computer-assisted interpreting tool using large language models. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Describes software application development rather than model compression or inference efficiency.
Ontology-Based Post-Coordinated Expression Assembly: from Clinical Narrative to Queryable Knowledge Graphs This paper introduces UNIGRAPHE, a deterministic method to assemble clinical text into queryable knowledge graphs conforming to SNOMED CT grammar. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Does not address model efficiency, quantization, distillation, or serving acceleration.
Communication, Vocalisation, and Language This position paper argues that large language models perform language as social practices rather than building distributed representations of symbols. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Does not address model compression, quantization, distillation, or efficient inference.
Know Yourself and Know Your Neighbour: A Syntactically Informed Self-Supervised Compositional Sentence Representation Learning Framework using a Recursive Hypernetwork This paper presents a self-supervised recursive hypernetwork framework using parse trees to learn syntactically informed compositional sentence representations. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Proposes a recursive hypernetwork architecture without addressing compression or serving efficiency.
SIREN: Situation-aware Index for Response and Emergency ActioNability in Crisis Translation This paper introduces an evaluation index to measure whether machine translation of emergency communication preserves critical safety actionability. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Assesses translation quality and impact rather than model compression or inference efficiency.
EHR-Bench: A Chinese Multi-Task Benchmark for Recruitment AI This paper introduces a Chinese multi-task evaluation benchmark covering six recruitment application areas for language models. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Benchmark paper comparing models, not developing small or efficient architectures.
Attention-Graph Curvature for the Verification of Mathematical Reasoning This paper evaluates whether attention-graph curvature features can help verifiers detect errors in mathematical reasoning steps. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Adds computational overhead to forward passes rather than improving model efficiency.
What Is the "Self" of a Large Language Model? This paper investigates internal model representations that separate assistant actions from externally provided text across conversational layers. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Examines geometric activation dynamics without focusing on compression or serving efficiency.
Learn from Move: the Next Step for GUI Agents This paper introduces an environment benchmark and exploration framework to help graphical user interface agents learn unfamiliar interaction dynamics. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Fine-tunes a 9B model but does not focus on efficiency, compression, or quantization.
Lost in Contrast: Instruction-Preserving Decoding for Reliable Visual Question Answering This paper introduces an instruction-preserving decoding strategy to prevent visual contrastive decoding from discarding valid abstention responses. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Investigates decoding contrast strategies rather than model compression or inference acceleration.
The Multilingual Artificial Hivemind: Do All LLMs Think Alike in Every Language? This paper investigates output diversity and text homogenization across twenty-three open-weight language models and thirteen languages. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not study model compression, quantization, distillation, or efficient inference.
Is the Platform Part of the Measurement? A Protocol and Simulation Study for Cross-Platform Equivalence of Machine-Labelled Policy Sentiment This paper presents a psychometric framework to test whether automated sentiment classification remains equivalent across different social media platforms. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not examine model compression, distillation, or efficient inference.
TPPO: Tracking-Value PPO For Language Model Reasoning This paper proposes a cumulative value parameterization method for reinforcement learning to improve credit assignment in multi-step language reasoning. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Focuses on RL value parameterization rather than model compression or inference efficiency.
The Bell Boundary of Shared Representations This paper uses Bell's theorem bounds to analyze capacity limits and task interference in multi-task shared representations. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Theoretical capacity bounds rather than practical distillation, pruning, or efficient inference.
Label Collapse in Diffusion Language Model In-Context Learning This paper analyzes label collapse failure modes across diffusion and autoregressive language models during in-context learning classification. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Studies diffusion and autoregressive model biases, not compression or serving efficiency.
ANCHOR: Assessing Naturalistic Constraint Hierarchies and Ordering in Repository-Grounded Agentic Coding This paper introduces a repository-grounded benchmark to evaluate how well coding agents follow engineering constraints and developer workflows. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Evaluates agent compliance on software tasks, not model compression or efficiency.
SpecGuardBench: Can Coding Agents Construct Reliable Validation Harnesses from Natural Language Specifications? This paper introduces a benchmark to test if coding agents can build reliable validation test harnesses from natural language specifications. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Focuses on evaluating agent testing capabilities, not model compression or efficiency.
Estimating Uncertainty of Omnimodal Large Language Models via External Omnimodal Embeddings This paper develops an uncertainty estimation framework for omnimodal models using external embeddings to score reasoning grounding and coherence. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not examine model compression, pruning, or efficient serving frameworks.
Lexical Redefinition Reveals Context-Sensitive Instruction Following This paper tests how language models handle explicit instructions when subsequent contextual sentences provide conflicting word definitions. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not investigate compression, quantization, distillation, or serving speed.
Event Memory and Narrative Dynamics: A Large Language Model Framework for Multidimensional State Spaces This paper presents a language model framework to extract event representations from autobiographical narratives into geometric state spaces. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model efficiency, quantization, or serving techniques.
Predicting Mental Health Help-Seeking Behaviour on Reddit: A Survival Analysis of Linguistic and Behavioural Signals This paper uses survival analysis on Reddit text and activity to predict when users will seek mental health support. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not study model compression, quantization, or inference optimization.
MENTOR: Procedural Shielding of Tool-Using LLM Agents This paper introduces a system to ensure tool-using language model agents comply with natural language company procedural manuals. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression, quantization, or inference speed.
AD4EDU: Benchmarking University Lecture Audio Description across Accessibility Standards, Metrics, and Learner Preferences This paper introduces a benchmark and rule-based metric for generating and evaluating audio descriptions in university slide lectures. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not focus on model efficiency, distillation, or edge deployment.
UZHCL_for_the_win at Touche: Encoder Fine-Tuning, Synthetic Augmentation, and an Underpowered Encoder-LLM Comparison This paper evaluates encoder fine-tuning and synthetic data augmentation for fallacy detection and argument classification on Reddit discussions. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Benchmarks standard encoders and LLMs without proposing compression or efficiency techniques.
Towards Universal Calibration for In-Context Classification This paper presents a universal calibrator to improve confidence estimates of in-context classification without requiring model-specific fine-tuning. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression, quantization, distillation, or serving speed.
Gluing Local Contexts into Global Meaning: A Sheaf-Theoretic Decomposition of Transformer Representations This paper proposes a mathematical framework using cellular sheaves to isolate phrasing-invariant content representations in transformer language models. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Examines internal representation geometry, not model compression or efficiency.
ToolHACE: A Dataset for Detecting Hallucinations in Tool-Based LLM Responses This paper introduces a benchmark dataset to evaluate hallucination detection in language model responses after external tool execution. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not focus on small models, quantization, distillation, or serving efficiency.
DuoDiscover: Scaling Autonomous Discovery Beyond the Evaluation Bottleneck The paper proposes an agent framework that discovers lightweight proxy evaluators to bypass costly benchmark evaluations during autonomous engineering discovery. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Addresses search and evaluation scaling rather than model compression or efficient serving.
Soft Teacher Forcing The paper proposes combining teacher-forced embeddings with soft tokens to improve supervised fine-tuning on reasoning datasets without changing inference. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Studies supervised fine-tuning training efficiency, not model compression or serving.
Welcome to the Age of Agent Ads: Using On-Policy Attacks to Steer Agentic Decisions The paper investigates on-policy indirect manipulation attacks that steer autonomous agent purchasing and ranking decisions via web retrieval. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression or inference serving efficiency.
Online LLM Personalization from Pairwise Preferences Across Users The paper formulates online language model persona personalization as a dueling bandit problem using sequential pairwise user preferences. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not focus on model compression, pruning, or serving efficiency.
Nufo: A Benchmark for Ewe Automatic Speech Recognition The paper introduces a 174-hour speech benchmark and evaluates baseline models on Ewe automatic speech recognition. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression or efficient inference techniques.
From Sampled Evidence to Radiology Reports: Objectives, Constraints, and Realization The paper evaluates how objective functions and candidate constraints affect decision conversion when generating radiology reports from sampled evidence. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not investigate model compression, quantization, or inference serving.
The Curse of Knowledge in Role-Playing Language Models The paper tests whether role-playing language models leak private or unobserved character information across simulated interactive scenes. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression or efficient serving.
UMCR: Universal Medical Concept Recognition through Guideline Integration and Agentic Linking The paper introduces a framework combining guidelines and agentic linking to ground medical concepts across diverse text types and ontologies. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not focus on model compression, quantization, or serving efficiency.
Evaluating Temporal Reasoning in Large Language Models: Event Temporal Relation Extraction with Allen's Interval Algebra, Logical Consistency, and Relative Time in Narratives The paper evaluates five language models on extracting complete interval networks of relative temporal events from narratives. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression or inference serving efficiency.
A Sense of Proprioception in Language Models The paper trains language models to report internal states through supervised fine-tuning and on-policy learning. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address inference acceleration, pruning, or model compression.
Everyone Was Right: Lexical Representations in Language Models Are Both Abstract and Holistic The paper investigates lexical representations in transformer models by decomposing representations into item, class, and interaction components. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Theoretical representation study, not addressing model compression or efficiency.
TDR: Tagalog Diacritic Restoration Dataset The paper introduces an annotated dataset and retrieval approach for Tagalog homograph diacritic restoration. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Mentions small models in passing, not focused on model efficiency techniques.
StateHarness: Maintaining Evolving Project State for Long-Horizon Coding Agents The paper presents a runtime framework that tracks evolving repository state and dependencies to assist long-horizon coding agents. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not study model compression, distillation, or inference optimization.
When Does an ASR Error Matter? Task-Role Diagnostics for Tool-Using Voice Agents The paper introduces a diagnostic protocol to evaluate how speech recognition errors affect structured tool calling in voice agents. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression, quantization, or efficient inference.
Decouple to Forget: Semantic Mediation for Sensitive Unlearning in Large Language Models The paper proposes an unlearning method that removes sensitive facts from model responses while preserving surrounding non-sensitive factual context. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Concerns knowledge erasure rather than model compression, quantization, or serving efficiency.
Worth Their Weights? Evaluating Generative Models for Discourse Parsing The paper compares discriminative and generative language models for rhetorical structure parsing and elementary discourse unit segmentation. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Focuses on parsing accuracy comparison across model sizes rather than compression techniques.
The Image Still Has a Say: Final-Answer Repair for Multimodal Chain-of-Thought The paper repairs multimodal chain-of-thought final answers using original image features without altering the generated reasoning trace. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression, quantization, or serving latency.
RALI@iKAT 2026: Retrieval-aware User Profile Denoising via Personalized Conversational Dense Retrieval The paper presents a retrieval-aware denoising method for conversational user profiles to improve personalized dense retrieval. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Focuses on retrieval relevance modeling rather than model compression or serving optimization.
Pushback in a Haystack: Language Models Respond to Factual Errors Ambiguously and Ineffectively The paper evaluates how language models correct user misconceptions across five languages, showing indirect pushback often fails to fix factual errors. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not cover model compression or inference acceleration.
Understanding Machine Taste: Aesthetic Agreement Across Foundation Models in K-pop The paper analyzes aesthetic judgments by foundation models on K-pop music videos to measure inter-model consensus across audio and visual modalities. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression, pruning, or serving speed.
Connecting the Dots: Enduring Performance Gaps in ASR Systems Between Standard and Minority English Varieties The paper tracks automatic speech recognition performance from 2021 to 2025, finding persistent accuracy disparities between standard American English and minority accents. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Focuses on accent fairness rather than model compression or inference efficiency.
Beyond Belief Nesting: Rethinking Higher-Order Theory of Mind in Large Language Models The paper introduces HopToM, a benchmark testing whether higher-order theory-of-mind tasks in language models genuinely evaluate nested reasoning rather than shortcuts. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Focuses on reasoning accuracy rather than model compression or inference efficiency.
GENDIS: Gender-Related Discourse and Stance in Multilingual YouTube Comments from India, Kenya, and Nigeria The paper presents GENDIS, a multilingual dataset of YouTube comments from India, Kenya, and Nigeria annotated for gender-related discourse and stance. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not focus on model compression, quantization, or efficient architectures.

1,621–1,680 of 2,430← Previous Next →