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 |
|---|---|---|---|
| Cross-Lingual TTS for Indian Languages: A Benchmark and Inference-Time Duration Estimation Strategies This paper introduces an Indian language text-to-speech benchmark and an inference-time duration estimation method for alignment-free synthesis. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Benchmark is for cross-lingual speech synthesis, not language model evaluation. | |
| Pre-Registered Genesis Accounting for Recursive Self-Improvement: First Audited Operator-Genesis Events in a Physics-Discovery Runtime This paper introduces an accounting framework to measure recursive self-improvement and audits machine-authored operator generation in a physics runtime. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents an accounting methodology for symbolic physics discovery, not LLM evaluation. | |
| Getting Space Right: Agentic Rewards for Image Generation This paper proposes a curriculum-guided reinforcement learning framework and an agentic reward model for spatial text-to-image generation. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on reward modeling and reinforcement learning for generative models, not LLM evaluation. | |
| PEAM: On-device Communication Agent for Personalized, Efficient and Autonomous Mobile Network Configuration This paper presents an on-device communication agent that uses small language models to generate low-latency mobile network configuration rules. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents an application and architecture rather than an evaluation benchmark. | |
| WeatherBridge: Forecast-Aware Multi-Source Input Adaptation for Pretrained Weather Forecasting Models This paper introduces an adapter that aligns operational numerical weather prediction inputs with ERA5 reference states for pretrained forecasting models. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on adapter architectures for weather forecasting, not LLM benchmarks. | |
| Beyond Outcome Feedback: Trajectory-Quality-Guided Experience-based Learning This paper presents an experience-based learning framework that uses trajectory quality signals rather than final task outcomes to guide agent improvement. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes an experience-learning mechanism rather than a benchmark or evaluation standard. | |
| MARL-BattleGrounds: A Novel JAX-native Benchmark for Heterogeneous and Competitive Multi-agent RL This paper introduces a JAX-native benchmark for evaluating heterogeneous and competitive multi-agent reinforcement learning algorithms. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Benchmark is for MARL game policies, not language model evaluation. | |
| Bridging Risk Approximation Gaps in Model Predictive Task Sampling via In-Context Modeling This paper addresses failure modes in model predictive task sampling for reinforcement learning by using in-context modeling. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on policy training sampling efficiency, not LLM evaluation frameworks. | |
| Predicting how model propensities generalize out of distribution This paper studies how language model behavioral propensities generalize out-of-distribution across different elicitation methods. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Studies behavioral propensity spillover rather than benchmarking or evaluation methodology. | |
| VC-OPD: Visual Counterfactual On-Policy Distillation for Grounded Vision-Language Reasoning This paper proposes visual counterfactual on-policy distillation to prioritize visually grounded tokens during vision-language reasoning training. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes a training method rather than an evaluation benchmark or methodology. | |
| Probabilistic Multi-Source Data Selection via Learnable Categorical Priors: Variance Reduction and Guaranteed Derandomization This paper introduces a probabilistic relaxation method with learnable categorical priors for bilevel multi-source data selection. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents optimization mathematics for data selection, not LLM benchmarking methods. | |
| Counterfactual Rollouts for Offline Contrastive Goal-Conditioned Reinforcement Learning This paper proposes a causal contrastive reinforcement learning method using counterfactual rollouts to handle unobserved confounding in offline demonstrations. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Theoretical RL algorithmic formulation without LLM evaluation or benchmark focus. | |
| MCPDisCo: Seesaw-Mitigated Tuning for MCP-Native Agents This paper presents a reinforcement learning framework to train language model agents across tools and servers in the Model Context Protocol ecosystem. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Introduces a training framework rather than a general LLM evaluation benchmark. | |
| What Survives When Activation Steering Is Used as Memory This paper evaluates the capabilities and limitations of using activation steering vectors as an agent memory mechanism. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Examines memory storage mechanisms rather than general LLM evaluation benchmarks. | |
| AlphaRLM: Aligning Language Models for Literature-Grounded Symbolic Factor Discovery This paper trains language models using reinforcement learning to construct executable financial trading formulas grounded in research literature. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on alpha discovery methodology rather than LLM evaluation benchmarks. | |
| FleetPilot: Fine-Grained Steering Empowers Real-World LLM Personalization This paper proposes a steering framework that dynamically personalizes language model responses using low-overhead activation vectors. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes an inference steering method rather than an evaluation methodology or benchmark. | |
| Single-Sample Decoding of a Two-Layer Sound-Word Canvas Model for Thai Klon Pad This paper introduces a two-layer masked language model approach to generate constrained Thai poetry in a single decoding pass. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on Thai poetry generation rather than standard LLM evaluation frameworks. | |
| Fitting Every Hint Is Not Enough: Premature Floods in Recurrent Algorithmic Reasoners This paper investigates failure modes where recurrent algorithmic reasoning models jump ahead prematurely when generalizing to larger graphs. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Analyzes failure modes in graph neural networks, not LLM evaluation benchmarks. | |
| Too Late to Prune: Generative AI, the Screen, and Team Creative Decisions This paper studies experimentally how the timing of generative AI access impacts team ideation and creative selection. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Investigates human organizational creativity rather than LLM benchmark design or evaluation. | |
| Mixing Regularization for Average-Reward $Q$-Learning and $Q$-Planning This paper provides theoretical convergence guarantees for average-reward reinforcement learning algorithms via mixing policy regularization. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Provides convergence theory for reinforcement learning rather than LLM evaluation. | |
| Certify the Decision, Not Every Score: Exact Top-k Selection from Few Evaluations This paper proposes an adaptive selection method that identifies the top candidates under uncertainty using minimal scenario evaluations. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Addresses scenario evaluation algorithms, not LLM testing benchmarks or LLM-as-a-judge. | |
| Recurrent In-Context Learning for Tabular Foundation Models This paper introduces parameter-tied recurrent blocks to reduce model size in tabular foundation models without sacrificing performance. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Introduces a model architecture rather than an evaluation framework or benchmark. | |
| On the Complexity Gap Between POMDPs and Multi-Environment POMDPs This paper maps the computational complexity landscape of robust decision-making in multi-environment partially observable environments. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Studies computational complexity theory, not language model benchmarks. | |
| UltraBERT: Beyond the Limit of BERT Pre-Training This paper analyzes pretraining factors such as masking ratios and token budgets to optimize bidirectional transformer encoders. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Investigates pre-training recipes rather than benchmarking frameworks or evaluation methodology. | |
| Toward System One Clinical Intelligence: Parameter-Efficient Fine-Tuning and Counterfactual Explainability of Decision Foundation Models This paper fine-tunes a quantized decision model to classify diseases from patient symptom descriptions without free-form text generation. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Evaluates on an existing benchmark rather than proposing new LLM evaluation methods. | |
| InCLAD: A Continual Learning Benchmark for Industrial Visual Anomaly Detection This paper introduces a benchmark to evaluate continual learning in semi-supervised visual industrial anomaly detection across components. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Benchmark is specifically for visual anomaly detection, not language model evaluation. | |
| Belief Embedding Tree Search This paper proposes a tree search method using belief embeddings for planning in partially observable environments. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Deals with reinforcement learning search algorithms rather than LLM benchmarking. | |
| Select or Rewrite: Oracle-Guided Fine-Tuning for Generative ASR Post-Editing from N-Best Hypotheses This paper fine-tunes language models to selectively correct speech recognition transcripts using curated candidate pools. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes a fine-tuning strategy rather than LLM benchmarking or evaluation tools. | |
| Deep Learning-based Semantic Segmentation of Dental Radiographs: Comparative Evaluation of PSPNet, DeepLabv3, UNet, and UNet++ This paper compares four deep learning segmentation models for detecting structures in dental radiographs under controlled experimental settings. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Applies standard CNN segmenters to dental images, not LLM evaluation benchmarks. | |
| Exposing AI Inference as a Telco Network API over the 5G User Plane This paper demonstrates an API and network framework to support split AI inference over 5G mobile networks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Covers telecommunications networking protocols, not LLM benchmark evaluation. | |
| GAMW: Generation-Aware Mel-Domain Watermarking for TTS Traceability This paper proposes a watermarking method embedded during speech synthesis to trace and verify generated audio. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Introduces watermarking techniques rather than LLM benchmarking or judge evaluations. | |
| Self-Evolving Visual Agent via Margin Reward This paper introduces a self-evolution framework that trains visual agents to invoke tools using reinforcement learning and unlabeled images. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on reinforcement learning for tool use rather than benchmark evaluation. | |
| Have You Been Convinced? A Neuro-symbolic Argumentation Reasoning Approach to Detecting Persuasiveness in ChangeMyView Discussions This paper presents a neuro-symbolic framework using language models and formal argumentation semantics to detect persuasiveness in online discussions. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Introduces an annotated discussion dataset rather than an LLM evaluation benchmark. | |
| VLCQ: Post-training quantization for deep neural networks using variable length coding Introduces a variable-length coding method for post-training neural network quantization that maintains accuracy without retraining. Peer-reviewed | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Addresses model compression algorithms, not benchmark contamination or LLM evaluation. | |
| TurnCost: Turns, Not Observations: Cache-Aware Loop Engineering for Coding Agents Analyzes coding-agent trajectories to model execution costs under prompt caching, showing turn count matters far more than observation token length. Peer-reviewed | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Studies trajectory costs and loops rather than presenting benchmark evaluation methodologies. | |
| Structural Chain-of-Thought: Reasoning with Executable Computational Graphs Replaces free-form reasoning text with executable computational graphs trained via reinforcement learning to improve reasoning speed and verifiability. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a structured reasoning technique rather than novel evaluation benchmarks. | |
| scHFBiMamba: a bidirectional Mamba autoencoder for cardiac disease-state classification from single-nucleus RNA-seq data Evaluates a compact bidirectional Mamba state-space autoencoder for classifying heart failure disease states from single-nucleus RNA sequencing data. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Addresses biological classification tasks, not general LLM evaluation benchmarks. | |
| ShellElastic: Scalable Elastic Pretraining with Simultaneous Maximal Updates and Hyperparameter Transfer Introduces an elastic transformer pretraining architecture enabling simultaneous hyperparameter transfer and subnetwork execution within a single model. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on architecture and hyperparameter transfer, not benchmark evaluation. | |
| RIECNN: Real-Time Image Enhanced CNN for Traffic Sign Recognition Presents an enhanced convolutional neural network architecture for real-time traffic sign classification in autonomous driving applications. Peer-reviewed | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Evaluates on standard traffic benchmarks rather than studying evaluation methodology. | |
| To Defer or Not to Defer? Online Learning with LLMs and Congested Experts Proposes an online learning algorithm that decides whether language models should answer queries directly or defer to congested human experts. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Studies human-in-the-loop deferral optimization, not language model benchmark testing. | |
| Trust Hinders Bidirectional Human-AI Alignment Critiques trust calibration in human-AI decision making for open-ended problems and outlines dimensions for higher-order system alignment. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents conceptual perspectives on trust rather than benchmark creation or validation. | |
| Single-step Retrosynthesis via Reinforcement Learning from Forward Feedback Trains chemical retrosynthesis models via reinforcement learning using forward reaction model consistency to credit plausible alternative pathways. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes a domain-specific model rather than general LLM evaluation frameworks. | |
| Harmony Harness: Multi-Objective Self-Evolution of Agent Harnesses Uses multi-objective evolutionary algorithms to optimize agent execution harnesses across competing metrics like accuracy and operating cost. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on optimizing agent harnesses rather than benchmarking methodology or dataset evaluation. | |
| Can agents whose sole duty is to keep humans alive be hijacked to kill? Delayed multi-turn derived-action trap attacks in simulated high-stakes environments Tests delayed prompt injection attacks on life-support control agents in a simulated lunar habitat to see if safety duties persist. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on safety and injection attacks rather than evaluating benchmark methodologies. | |
| Continuous Risk-Based Scheduling for Food Safety Inspection: Integrating Historical Decay with Event-Driven Consumer Review Signals Combines historical violation decay with language-model triage of online consumer reviews to dynamically schedule restaurant health inspections. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Focuses on risk scoring algorithms rather than LLM evaluation benchmarks. | |
| SocraticRepair: Planning-Guided Socratic Tutoring for Program Debugging and Misconception Repair Fine-tunes a coding model using synthetic tutoring dialogues to provide step-by-step guidance without revealing program fixes prematurely. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes an educational tutoring system rather than an evaluation benchmark. | |
| Exploring Jailbreak Vulnerability Across Depth-Based Dynamic Inference in Large Language Models Studies how depth-based early-exit inference affects language model safety, finding shallow exits resist jailbreaks better than deep exits. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Analyzes vulnerability properties rather than presenting an LLM benchmark. | |
| PANINI-DR: Controlled Deep Research through Generative Semantic Workspaces Presents a multi-agent framework that decouples research planning, subquestion investigation, and factual verification via a shared semantic workspace. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a deep-research multi-agent framework rather than an evaluation benchmark. | |
| Running Fewer Loops Removes Refusal Behaviour in Recurrent-Depth Language Models Shows that running fewer layer loops in recurrent-depth models significantly reduces safety refusals while preserving most reasoning performance. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Investigates internal refusal mechanisms rather than proposing an LLM benchmark. | |
| HarnessNet: Feedback-driven Hierarchical Joint Harness Evolution for Multi-agent Systems Proposes a hierarchical evolution framework that allows individual agent software harnesses in multi-agent systems to adapt using feedback. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes a multi-agent system optimization framework, not an evaluation benchmark. | |
| Understanding Unified VLMs: From Attention Probing to Guidance Tuning for Multimodal Generation Probes attention heads in unified vision-language models to study condition processing and introduces an inference-time guidance tuning method. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents an inference guidance technique rather than an evaluation benchmark. | |
| Using Evidence-Centered Design to Guide Prompt Engineering in Measurement: An Illustration with Automated Essay Scoring Applies an educational measurement framework to prompt engineering to improve the validity and reproducibility of automated essay scoring systems. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Covers prompt engineering methodologies for scoring rather than LLM evaluation benchmarks. | |
| MoMaTune: Model, Match, Then Tune Language Models for Numerical Records Presents a two-stage post-training strategy that improves language model handling of tabular numerical records using auxiliary pretraining and token matching. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents a fine-tuning methodology rather than evaluation methodologies or benchmarks. | |
| Framework for an AI-Ready ARPES Community Database Proposes a data-sharing ecosystem and framework to make angle-resolved photoemission spectroscopy data organized and accessible for AI research. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Discusses experimental data sharing rather than LLM benchmarking. | |
| EvolvAttack: Backdooring self Evolving VLM Agents The paper introduces EvolvAttack, demonstrating a backdoor mechanism that activates only after a vision-language agent evolves its memory over time. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Studies backdoor attack vectors rather than general model evaluation methods. | |
| Sequent-Prover: Training Agents for Formal, Checkable SMT-based Reasoning The paper introduces Sequent-Prover, training language model agents to perform multi-turn formal reasoning using an external SMT solver scratchpad. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Introduces an agent framework and training method, not a general benchmark. | |
| RECAP: Calibrating Low-Bit Reasoning Models on the History They Read The paper presents RECAP, a post-training quantization method that calibrates reasoning models using dynamic prefixes to reduce error accumulation. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes a post-training quantization method rather than benchmark evaluation. | |
| LeHami: Symplectic Latent Dynamics for World Models The paper proposes LeHami, incorporating symplectic Hamiltonian dynamics into latent predictive architectures to prevent representation collapse over long rollouts. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Proposes a world model architecture, not an evaluation framework. | |
| GeoBridge: Evolving Transferable Decoder Topologies for Geometric Vision The paper presents GeoBridge, an evolutionary framework that searches for reusable, transferable decoder topologies across geometric vision tasks. Preprint | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Presents an architectural search framework, not an LLM benchmark. | |
| Protein Crystallization Segmentation and Classification Using Subordinate Color Channel in Fluorescence Microscopy Images The paper proposes using subordinate color channel intensities in fluorescence microscopy to improve protein crystallization segmentation and classification. Peer-reviewed | LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17 | Not relevant Addresses microscopy image segmentation rather than LLM benchmarking or evaluation. |
2,221–2,280 of 3,345← PreviousNext →