Research Feed Page 24

The paper investigates how post-training quantization and activation calibration impact cross-sectional volatility forecasting models applied to financial time series.

The paper tracks massive activation tokens across hybrid linear attention large language models to understand how layerwise hybridization reshapes internal activation dynamics.

The paper introduces an evidence-grounded framework using large language models to support medical-device safety analysis across the product lifecycle.

The paper introduces NetlistBench, a benchmark containing 2,342 cases across 24 task families to evaluate large language model reliability in recognizing and manipulating simulator-facing SPICE netlists.

The paper introduces Preference Tree Optimization, a method that uses look-ahead simulations to enhance goal-oriented dialogue systems by training models to prefer paths that lead to better future conversation outcomes.

The paper introduces RT-SEMamba, a fully causal, Mamba-based speech enhancement architecture operating online with a 25 ms algorithmic latency constraint via progressive knowledge distillation.

The paper introduces Behavioral Mode Axes to reliably measure and steer large language model behavioral styles at inference time, overcoming the limitations of unstable self-report questionnaires.

This paper investigates how prompt-level specialist decomposition and task-aligned reinforcement learning improve financial analysis in European listed real estate.

The researchers developed a new benchmark and multimodal agent system that uses visual data to generate more realistic business ideas than text-only alternatives.

DreamFly introduces a memory-augmented diffusion architecture that uses receding horizon planning to navigate complex aerial environments.

Map-Det3D uses a metric 3D reconstruction model to improve the accuracy of detecting objects in 3D space from streamed video inputs.

Self-Geometry is a plug and play pipeline that improves the geometric consistency of pretrained 3D vision foundation models during inference without needing ground truth data.

This paper addresses simulator collapse in multi-agent reinforcement learning by introducing techniques to diversify simulator responses and jointly train policies with multiple simulators.

Mechanist is a multi-agent system that autonomously explores AI model internals to discover mechanistic theories about how these models function.

The paper introduces Diagram-MMU, a benchmark evaluating how effectively Multimodal Large Language Models handle scientific diagram parsing, editing, and question answering.

The paper introduces a curvature aware zeroth order optimization method that reduces memory usage during test time model adaptation without requiring backpropagation.

The AVA-Encoder converts complex video content into a structured knowledge graph to enable more accurate video reconstruction and agent-based editing.

The researchers introduced CTBench, a new evaluation framework designed to measure how effectively AI agents diagnose and resolve issues in complex, heterogeneous telecommunications network environments.

The paper introduces an agentic framework that uses multimodal large language models and Bayesian optimization to align Image-to-Video model outputs with creative briefs, replacing trial-and-error workflows.

The authors implement a Deep Q-Network to provide real-time, autonomous intrusion detection and threat mitigation for cloud infrastructure.