Every paper checked — nothing hidden
All papers
Every paper checked in the week of 12–18 Oct 2026 · LLM evaluation and benchmarks — nothing is hidden. Brief papers first.
All3,345In brief3More (on-topic)587Not relevant2,721Not assessed24Older45Could not confirm25Blogs with no paper54All blogs57
| Paper | Topic | Relevance | Feedback |
|---|---|---|---|
| Agent Sociodrama: Social-Dynamics-Guided Persona Recasting for LLM Agent Teams The paper introduces a framework to model evolving social dynamics and update personas in multi-agent language model teams. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents an agent coordination method rather than an evaluation benchmark. | |
| Adaptive Look, Flexible Thought: Learning How and How Long to Think in Multimodal Reasoning The paper proposes a reasoning framework that discretizes visual states into tokens to jointly optimize visual computation and chain-of-thought steps. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a multimodal reasoning framework rather than general LLM evaluation. | |
| LLM Value Functions for Credit Assignment in Agentic RL The paper uses in-context language model evaluations as state value functions to improve credit assignment in agentic reinforcement learning. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a reinforcement learning training method rather than an evaluation benchmark. | |
| Attention Sinks in VLMs Are Borrowed, Not Seen The paper analyzes visual attention sinks in vision-language models and shows they stem from alignment with language text sinks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on internal model attention mechanics rather than evaluation benchmarks. | |
| Personalized LLM Recommendation with Semantic IDs: Autoregressive Beam Search vs. Multi-Token Prediction The paper compares single-head autoregressive beam search with parallel multi-token prediction for generating semantic IDs in recommendation systems. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Compares two decoding objectives rather than general LLM evaluation benchmarks. | |
| Engram Injection: A Study of Temporal Validity in Parametric Language-Model Memory The paper studies whether language models can track temporal validity changes in parametric memory via adapter conditioning and clocks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Investigates parametric memory updates rather than LLM benchmarking methods. | |
| What the Model Cannot See: Covert and Provable Text Invisibility in LLMs The paper examines how trigger-wrapped text can be made completely invisible to language model activations and downstream tasks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on security backdoor attacks rather than benchmark methodology. | |
| Comparative Study of Canvas- and Coordinate-Based Geometry-Conditioned Neural Networks for Real-Time Prediction of Indoor Environments with Varying Room Dimensions The paper compares canvas-based and coordinate-based neural network surrogates for predicting indoor airflow and temperature fields under varying room dimensions. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Addresses CFD surrogate modeling, not general AI benchmark evaluation. | |
| Variance-Optimal State Resampling for Reinforcement Learning with Verifiable Rewards The paper proposes variance-driven state resampling to optimize the state sampling distribution for reinforcement learning with verifiable rewards. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on RL optimization algorithms rather than benchmark design or testing. | |
| Learning Multimodal Sensor Fusion with JEPA Representations for Breathing Activity and Pose Recognition The paper explores transferring frozen video representations across heterogeneous sensor subsets to classify breathing activity and pose. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Evaluates human sensor classification rather than general LLM benchmarking. | |
| Make Every Expansion Count: Learning to Search Efficiently in LLM Code Repair The paper introduces a learned search policy that predicts required repair steps to prioritize code edits before generating and testing patches. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on optimizing code search rather than LLM benchmarking methodologies. | |
| Gen Through the Lens: Camera-Controllable Video Generation via Cross-Coordinate Conditioning Fusion The paper proposes an image-to-video framework combining three coordinate spaces to enable camera-controllable video generation. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Does not evaluate language models or general AI benchmarks. | |
| LURE: a Phase-Labelled Synthetic Corpus of Online-Grooming Exchanges The paper presents a phase-annotated synthetic conversational corpus generated using a retrieval-augmented pipeline to study online grooming of minors. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a domain-specific dataset for criminology rather than LLM benchmarking. | |
| RINO: Renormalization Group Distillation with No Labels The paper presents an unsupervised pretraining method that uses the multiscale structure of jet formation for high-energy physics data. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Does not address language model benchmarking or evaluation. | |
| Multi-Horizon Explainable Queue Forecasting and Resource-Aware Decision Support for Intelligent Border Control Using LightGBM The paper presents a LightGBM architecture for multi-horizon queue forecasting and resource-aware decision support in border control systems. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on border queue prediction, not LLM benchmark evaluation. | |
| DiffNash: Finding Nash Equilibrium in High-Dimensional Continuous Action Spaces with Diffusion Policies The paper presents a diffusion-based framework that uses reweighted score matching to compute Nash equilibria in continuous multi-agent action spaces. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Does not evaluate LLMs or introduce general evaluation frameworks. | |
| Vision: "What Don't I Know About User's Preferences?" How Uncertainty Estimation Improves Proactive Assistant Alignment The paper argues that uncertainty estimation in smart assistants is necessary to prevent unsafe actions when user preferences remain ambiguous. Peer-reviewed | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Is a vision paper on uncertainty, not a benchmark or LLM evaluation study. | |
| Hardware-Assisted Multi-Stage Ransomware Detection with Adaptive Machine Learning The paper combines static software scanning with hardware-level microprocessor monitoring and neural architecture search to detect ransomware in real time. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Does not address LLM evaluation methods or language benchmarks. | |
| ADAPT: Audio-Depth Action Planning with Transition-Aware Rewards for Lights-Out Industrial Navigation The paper introduces a navigation framework and benchmark using depth and spatial audio for obstacle avoidance in unlit industrial environments. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a benchmark for navigation robotics, not broader language model evaluation. | |
| Unfair and Unsafe: Everyday Bias Triggers Emergent Misalignment The paper finds that fine-tuning language models on polite text containing demographic stereotypes triggers broader safety and truthfulness failures on unrelated tasks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Investigates safety degradation from fine-tuning rather than benchmark design or judge reliability. | |
| Small but Mighty: Test Suites as Improvement Signals for Self-Evolving Agents The paper presents an evolutionary search method that generates small, high-quality test suites to verify code rewrites in self-evolving agents. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on agent self-improvement search rather than general evaluation benchmarks. | |
| PopGen: Generating Semantically Diverse Populations for Zero-Shot Coordination The paper uses language models to generate semantically diverse agent policies in code to train reinforcement learning cooperators for zero-shot collaboration. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on population-play training for reinforcement learning, not LLM evaluation benchmarks. | |
| Embedding Information in Policy Randomness: Distribution-Preserving Implicit Communication in MDPs The paper presents a coding method that embeds hidden communication into an agent's policy randomness without altering its observable action distribution. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Theoretical channel coding paper, not an LLM benchmark evaluation study. | |
| Revocation Is Not Complete at Receipt: An Executable Outcome Contract for Distributed Agent-Tool Gateways The paper introduces an executable contract and monitor to verify that delegated authority revocations actually take effect across distributed agent gateways. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Designs distributed systems contracts, not LLM capability benchmarks. | |
| Handoff Curves: Initial Selection, Later Handoff, and Supervision Targets in Multi-turn LLM Routing The paper evaluates model handoff points in multi-turn routing workflows, analyzing trade-offs between initial model selection, mid-task switching, and operational cost. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Studies routing optimization metrics rather than general LLM evaluation methods. | |
| AutoSec-RAN: Specification-Grounded LLM Fine-Tuning for Automated Security Analysis of O-RAN The paper fine-tunes a language model on technical telecommunication specifications to automate security incident triage in Open Radio Access Networks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Domain-specific security triage benchmark rather than general LLM evaluation. | |
| Computational Reflective Equilibrium for Agentic AI: A Rawlsian Alignment Scaffold for Governance The paper proposes an ethical governance scaffold based on reflective equilibrium to resolve conflicting instructions and policy rules in agentic workflows. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on governance scaffolding frameworks rather than benchmark testing. | |
| When Does Compensating Inherited Quantization Error Help? Mean-Square Guarantees for Layer-Wise Quantization and Their Limits The paper provides theoretical mean-square error guarantees and bounds for layer-wise quantization error compensation across linear and non-linear networks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Analyzes post-training quantization error bounds, not LLM benchmark methodologies. | |
| Of Errors and Echoes: Responsibility-Aware Credit Assignment for Tool-Use Agents The paper introduces an optimization method that separates error identification from credit assignment to improve reinforcement learning for tool-using agents. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on agent credit assignment optimization rather than benchmark evaluation. | |
| OT-FlowRec: Optimal Transport-Guided Multi-Bridge Flow for Multimodal Sequential Recommendation The paper presents an optimal transport flow-matching framework to model concurrent, multi-interest user behavior in multimodal sequential recommendation systems. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Concerns recommender system flows rather than language model evaluation frameworks. | |
| When Agents Talk, Secrets Travel: Topology-Driven Leakage and Recipient-Conditioned Sanitization in Multi-Agent LLM Systems The paper investigates privacy leakage caused by multi-agent network topologies and proposes an execution sanitizer conditioned on recipient identity. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on data leakage sanitization rather than benchmark validity testing. | |
| Autoregressive Decoding as a Martingale: Characterizing the Stochastic Dynamics of LLM Generation The paper models autoregressive decoding errors as a martingale process to analyze the stochastic dynamics of language model generation. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Analyzes martingale decoding dynamics, not benchmark design or contamination. | |
| AI Representations of Mental Health Stigma: Quantifying the Gap Between AI Estimations and Human Perceptions Across Gender Groups Using GPT-Based LLMs The paper compares mental health stigma representations in large language models against human survey data across different gender groups. Peer-reviewed | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Measures gender stigma representation rather than general LLM benchmarking methods. | |
| Adaptive Compression and Targeted Perturbation: A Unified Framework for Generalized Audio Deepfake Detection The paper introduces an audio deepfake detection framework that dynamically selects latent dimensions and uses spectrogram perturbations to improve generalization across attacks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Targets audio spoofing detection rather than LLM benchmarking methods. | |
| Cross-Protocol Personality Assessment with LLMs: Aligning Behavioral Evidence with Human Ratings The paper proposes a framework to transfer personality assessment across distinct asynchronous video interview protocols using construct alignment and language model raters. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Studies cross-protocol personality scoring rather than foundational LLM evaluation standards. | |
| Understanding and Reducing Misalignment in Persuasion Optimization The paper studies how reinforcement learning for persuasion causes models to fabricate evidence and coercive tactics, evaluating strategies to reduce this behavior. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Evaluates persuasion alignment dynamics rather than general LLM benchmark reliability. | |
| Dream2Discern: Learning from Hard Latent Variants for Runtime Failure Prediction in Multi-Agent Systems The paper presents a runtime auditor that predicts execution failures in multi-agent systems using loss-guided perturbations to identify hard failure cases. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Introduces a runtime auditor model rather than benchmark measurement methodology. | |
| Beyond Single-Pathogen Forecasts: Predicting Shared Respiratory Hospital Burden from Influenza, COVID-19, and RSV This paper evaluates methods for forecasting total hospital respiratory burden by comparing direct combined prediction against summing individual pathogen models. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Studies hospital burden forecasting metrics, not language model benchmark methodologies. | |
| Heterogeneous Agentic Coevolution This paper proposes a coevolution framework where language model agents and domain foundation models iteratively train and improve each other. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focus is on agent co-training, not LLM benchmark evaluation. | |
| The Sharp Directions Are Against You: Curvature Analysis of Activation Steering This paper uses loss curvature to identify and remove fragile directions that cause activation steering vectors to fail. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on internal representation analysis rather than standard benchmarks. | |
| Where Polanyi Meets Galois: Tacit Knowing as an Algebro-Deterministic Operation, Toward Re-rooting AI-Generated Science This paper models tacit versus explicit knowledge in scientific discovery as distinct operational regimes in content-addressable memory hardware. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Hardware architecture theory, not an LLM evaluation benchmark. | |
| GUIDE: Computationally Augmented Consensus for Trustworthy Clinical Practice Guidelines This paper introduces GUIDE, a multi-agent copilot system that assists clinicians in drafting and updating evidence-based medical practice guidelines. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a clinical system rather than a general LLM benchmark. | |
| Constructed regional photovoltaic generation fields improve ultra-short-term distributed solar forecasting This paper develops SatPV-Field to forecast distributed solar generation by converting fleet measurements into spatial fields with satellite guidance. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Application-specific forecasting framework, not an LLM benchmark. | |
| Imitation Holds a Key to Evolution: Localized Repair and Verified Skill Acquisition for Adaptive Agents This paper presents an imitation framework that enables agents to identify violated contracts, extract skills, and repair execution failures. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes an adaptive repair controller rather than an evaluation benchmark. | |
| PID-Gated Continual Prompt Adaptation Under Noisy Feedback This paper proposes control-inspired gates to decide when language model agents should adapt their prompts without absorbing noisy user feedback. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes an adaptation mechanism rather than an evaluation benchmark. | |
| Omni-Modal LLMs Know More Than They Say This paper analyzes why omni-modal language models perceive non-textual inputs correctly yet fail to mention them in text outputs. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on alignment methodology rather than evaluation benchmarks. | |
| Large Language Model Assisted Transportation Planning through a Differentiable User Equilibrium Layer This paper combines language models with a differentiable traffic equilibrium layer to help non-experts optimize transportation planning problems. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Describes an optimization application rather than an evaluation benchmark. | |
| Phantom Merge: When Your Large Language Model Agents Pick One but Tell You About Another This paper identifies and mitigates Phantom Merge, an error where language model agents bind retrieved facts to the wrong entity. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on error analysis and mitigation rather than a benchmark. | |
| Pollard: Governed Semantic Execution Trees for Framework-Independent AI Agent Control and Replay This paper presents Pollard, a framework-independent agent runtime that uses semantic execution trees for deterministic tracking, governance, and replay. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on agent runtime infrastructure, not evaluation benchmarks. | |
| Unravel: Recovering an Interactive 3D Scene from a Single Image This paper introduces Unravel, a training-free framework that converts a single image into an editable, multi-object 3D scene. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a 3D reconstruction system, not an evaluation benchmark. | |
| Fisher information and the geometry of memorization in neural networks This paper analyzes Fisher information eigenvectors to show how neural networks geometrically decouple generalizable patterns from memorized noisy data. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Theoretical analysis of memorization, not an evaluation benchmark. | |
| Interactive 4D Volumetric Liquid Forecasting under Moving-Solid Interaction This paper proposes a task and framework to forecast dense 3D liquid velocity fields based on how solid objects move through them. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a physics modeling task, not an LLM benchmark. | |
| EventLens: Event-Structure Reinforcement Learning for Video Understanding This paper presents EventLens, a reinforcement learning framework that trains multimodal models to understand video through explicit event structure recovery. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Introduces a reinforcement learning framework, not a general evaluation benchmark. | |
| ProCARE: Real-World Study Automation via Profile-Grounded Evidence Contracts This paper proposes ProCARE, a framework that uses structured contracts to reliably automate healthcare observational studies with language model agents. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on execution framework rather than general LLM evaluation. | |
| Label-Centric Knowledge Retrieval for Zero-Shot Text Classification This paper proposes a framework that precomputes label definitions and boundaries with large models to enable cheaper zero-shot text classification. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a classification method, not an evaluation benchmark. | |
| MP-MNCA: Morphology-Prior Masked Neighbor Cross-Attention for Efficient Spatial Domain Identification This paper presents a cross-attention method with histological priors to improve training efficiency and convergence in spatial transcriptomics. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Biology application model, not an LLM evaluation method. | |
| SafeDSE: Joint Exploration of Safety Architecture Patterns, Mapping, and Scheduling for Dataflow Applications With Safety Arguments (Extended Version) Presents a design space exploration tool for mapping, scheduling, and generating safety arguments for dataflow applications on embedded hardware platforms. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Embedded hardware exploration framework, not an LLM benchmark. | |
| OneShot: Label-Efficient Detector Weighting for Time Series Anomaly Detection Weights an ensemble of time-series anomaly detectors using confirmation of a single reported anomaly from a human operator. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes a detector ensembling method, not an LLM benchmark. | |
| Look Back to Move Forward: Learning to Revisit for Agentic Compositional Image Generation Develops an agentic text-to-image framework allowing agents to revisit and select earlier image states for iterative compositional generation. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes an agent framework, not a benchmark. | |
| AutoNovo: Agentic Program Search Discovers Transferable Nucleic Acid Design Algorithms Uses agentic program search to iteratively write and refine design algorithms for nucleic acid sequences that transfer across independent models. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes a design method, not an LLM benchmark. |
2,341–2,400 of 3,345← PreviousNext →