3/10/2026
Targeted Speaker Poisoning Framework in Zero-Shot Text-to-Speech
Speech Generation Speaker Poisoning framework for removing specific speaker identities from zero-shot TTS models via machine unlearning.
3/10/2026
Nw\=ach\=a Mun\=a: A Devanagari Speech Corpus and Proximal Transfer Benchmark for Nepal Bhasha ASR
Nwachā Munā: 5.39-hour manually transcribed Devanagari corpus for endangered Nepal Bhasha with ASR benchmark and cross-lingual transfer.
3/10/2026
GRD-Net: Generative-Reconstructive-Discriminative Anomaly Detection with Region of Interest Attention Module
GRD-Net: generative-reconstructive-discriminative model with attention for surface defect detection and localization.
3/10/2026
A Systematic Comparison of Training Objectives for Out-of-Distribution Detection in Image Classification
Systematic comparison of 4 training objectives for OOD detection: Cross-Entropy, Prototype, Triplet, and AP Loss.
3/10/2026
SMAT: Staged Multi-Agent Training for Co-Adaptive Exoskeleton Control
SMAT: four-stage curriculum learning for co-adaptive exoskeleton control modeling sequential human motor adaptation.
3/10/2026
AtomicVLA: Unlocking the Potential of Atomic Skill Learning in Robots
AtomicVLA: Visual-Language-Action model for multi-step robotic manipulation with atomic skill learning and continual adaptation.
3/10/2026
TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable Reward
TDM-R1: reinforcement learning framework for few-step diffusion models using non-differentiable rewards like human feedback.
3/10/2026
VoiceSHIELD-Small: Real-Time Malicious Speech Detection and Transcription
VoiceSHIELD-Small: lightweight real-time model for detecting malicious speech and prompt injection in voice interfaces without speech-to-text delay.
3/10/2026
YAQIN: Culturally Sensitive, Agentic AI for Mental Healthcare Support Among Muslim Women in the UK
YAQIN: culturally-sensitive agentic AI application for mental healthcare support integrating Islamic frameworks for Muslim women in UK.
3/10/2026
QuadAI at SemEval-2026 Task 3: Ensemble Learning of Hybrid RoBERTa and LLMs for Dimensional Aspect-Based Sentiment Analysis
Ensemble approach combining RoBERTa and LLMs for aspect-based sentiment analysis in SemEval-2026 task via prediction-level fusion.
3/10/2026
ProgAgent:A Continual RL Agent with Progress-Aware Rewards
ProgAgent: continual RL agent using progress-aware reward learning from expert videos for lifelong robotic learning, addressing catastrophic forgetting.
3/10/2026
Dual-Metric Evaluation of Social Bias in Large Language Models: Evidence from an Underrepresented Nepali Cultural Context
Evaluates social and cultural biases in 7 LLMs (GPT-4o-mini, Claude, Gemini, Llama, Mistral) using Nepali cultural context with dual-metric methodology.
3/10/2026
HybridStitch: Pixel and Timestep Level Model Stitching for Diffusion Acceleration
Method for accelerating diffusion-based text-to-image generation through pixel and timestep-level model stitching.
3/10/2026
Gradient Iterated Temporal-Difference Learning
Theoretical advancement in temporal-difference learning for reinforcement learning addressing divergence issues in gradient TD methods.
3/10/2026
AI Misuse in Education Is a Measurement Problem: Toward a Learning Visibility Framework
Proposes measurement framework for addressing AI misuse in education rather than detection-based approaches to academic integrity.
3/10/2026
DistillGuard: Evaluating Defenses Against LLM Knowledge Distillation
Framework systematically evaluating defenses against knowledge distillation attacks on proprietary LLM APIs with taxonomy of mitigation strategies.
3/10/2026
AI Steerability 360: A Toolkit for Steering Large Language Models
Open-source toolkit for steering LLMs through four control surfaces: input, structural, state, and output modifications.
3/10/2026
CCR-Bench: A Comprehensive Benchmark for Evaluating LLMs on Complex Constraints, Control Flows, and Real-World Cases
Comprehensive benchmark for evaluating LLMs on complex instruction following with intricate control flows and real-world constraints.
3/10/2026
Reject, Resample, Repeat: Understanding Parallel Reasoning in Language Model Inference
Theoretical analysis of inference-time sampling aggregation in LLMs using particle filtering lens to understand accuracy-cost tradeoffs.
3/10/2026
VLM-SubtleBench: How Far Are VLMs from Human-Level Subtle Comparative Reasoning?
Benchmark dataset evaluating vision-language models on subtle visual differences for anomaly detection and medical imaging tasks.
3/10/2026
Long-Short Term Agents for Pure-Vision Bronchoscopy Robotic Autonomy
Vision-only autonomy framework for robotic bronchoscopy using long-short term agent architecture for intraoperative navigation.
3/10/2026
IMSE: Intrinsic Mixture of Spectral Experts Fine-tuning for Test-Time Adaptation
Test-time adaptation method using spectral experts in Vision Transformers for improved performance on distribution-shifted data.
3/10/2026
SWE-Fuse: Empowering Software Agents via Issue-free Trajectory Learning and Entropy-aware RLVR Training
LLM-based software agent framework using trajectory learning and entropy-aware reinforcement learning for issue fixing tasks.
3/10/2026
AI Agents, Language, Deep Learning and the Next Revolution in Science
Position paper proposing AI agents supervised by humans to address data-understanding imbalance across scientific disciplines.
3/10/2026
ELLMob: Event-Driven Human Mobility Generation with Self-Aligned LLM Framework
LLM-based framework for generating human mobility trajectories during large-scale events using event-annotated datasets.
3/10/2026
Emergence is Overrated: AGI as an Archipelago of Experts
Research paper arguing that intelligence requires efficient representations and generalization rather than emergent capabilities from large model assemblies.
3/10/2026
\$OneMillion-Bench: How Far are Language Agents from Human Experts?
$OneMillion-Bench: benchmark with 400 expert-curated tasks across law, finance, healthcare evaluating language agents on real-world professional demands.
3/10/2026
ViSA-Enhanced Aerial VLN: A Visual-Spatial Reasoning Enhanced Framework for Aerial Vision-Language Navigation
ViSA-Enhanced Aerial VLN: visual-spatial reasoning framework for aerial vision-language navigation with triple-phase collaboration.
3/10/2026
FedMomentum: Preserving LoRA Training Momentum in Federated Fine-Tuning
FedMomentum: federated fine-tuning method for LLMs with LoRA that preserves training momentum via improved aggregation.
3/10/2026
Alignment--Process--Outcome: Rethinking How AIs and Humans Collaborate
Framework reconceptualizing AI-human collaboration through alignment, process structure, and outcome quality dimensions.
3/10/2026
Not Like Transformers: Drop the Beat Representation for Dance Generation with Mamba-Based Diffusion Model
MambaDance: Mamba-based diffusion model for music-synchronized dance generation emphasizing sequential and rhythmical characteristics.
3/10/2026
DyLLM: Efficient Diffusion LLM Inference via Saliency-based Token Selection and Partial Attention
DyLLM: efficient diffusion language model inference using saliency-based token selection and partial attention mechanisms.
3/10/2026
Solution to the 10th ABAW Expression Recognition Challenge: A Robust Multimodal Framework with Safe Cross-Attention and Modality Dropout
Multimodal framework with safe cross-attention for emotion/expression recognition handling occlusions and missing modalities.
3/10/2026
Speed3R: Sparse Feed-forward 3D Reconstruction Models
Speed3R: sparse feed-forward 3D reconstruction model reducing computational complexity from dense attention mechanisms.
3/10/2026
ImageEdit-R1: Boosting Multi-Agent Image Editing via Reinforcement Learning
ImageEdit-R1: multi-agent image editing system enhanced with reinforcement learning to handle complex multi-step user instructions.
3/10/2026
DSH-Bench: A Difficulty- and Scenario-Aware Benchmark with Hierarchical Subject Taxonomy for Subject-Driven Text-to-Image Generation
DSH-Bench: comprehensive benchmark with hierarchical taxonomy for evaluating subject-driven text-to-image generation models.
3/10/2026
DC-W2S: Dual-Consensus Weak-to-Strong Training for Reliable Process Reward Modeling in Biological Reasoning
DC-W2S method for training reliable process reward models in biological reasoning tasks using weak supervision and dual consensus.
3/10/2026
SaiVLA-0: Cerebrum--Pons--Cerebellum Tripartite Architecture for Compute-Aware Vision-Language-Action
SaiVLA-0: neuroscience-inspired vision-language-action architecture with tripartite design for compute-aware embodied AI tasks.
3/10/2026
Foley-Flow: Coordinated Video-to-Audio Generation with Masked Audio-Visual Alignment and Dynamic Conditional Flows
Foley-Flow model for coordinated video-to-audio generation using masked audio-visual alignment and conditional flows.
3/10/2026
DARC: Disagreement-Aware Alignment via Risk-Constrained Decoding
DARC method for inference-time LLM alignment that handles heterogeneous human preferences via risk-constrained decoding without retraining.
3/10/2026
Gradually Excavating External Knowledge for Implicit Complex Question Answering
Framework for open-domain question answering that gradually excavates external knowledge to overcome LLM limitations in implicit reasoning tasks.
3/10/2026
An explainable hybrid deep learning-enabled intelligent fault detection and diagnosis approach for automotive software systems validation
Explainable deep learning approach for fault detection and diagnosis in automotive software validation during V-development process.
3/10/2026
Evolution Strategy-Based Calibration for Low-Bit Quantization of Speech Models
Evolution strategy-based calibration method for quantizing speech processing models to address audio-specific challenges in low-bit quantization.
3/10/2026
Is continuous CoT better suited for multi-lingual reasoning?
Research on continuous chain-of-thought reasoning in latent space for multilingual capabilities across 5 languages using GSM8k and CommonsenseQA datasets.
3/10/2026
Privacy-Preserving End-to-End Full-Duplex Speech Dialogue Models
Analysis of speaker privacy leakage in end-to-end full-duplex speech dialogue LLM models across layers.
3/10/2026
TildeOpen LLM: Leveraging Curriculum Learning to Achieve Equitable Language Representation
30B-parameter open-weight LLM for 34 European languages using curriculum learning for linguistic equity.
3/10/2026
MM-TS: Multi-Modal Temperature and Margin Schedules for Contrastive Learning with Long-Tail Data
Multi-modal temperature and margin schedules for contrastive learning with imbalanced data.
3/10/2026
Distributional Regression with Tabular Foundation Models: Evaluating Probabilistic Predictions via Proper Scoring Rules
Evaluation framework for tabular foundation models using distributional regression and proper scoring rules.
3/10/2026
SplitAgent: A Privacy-Preserving Distributed Architecture for Enterprise-Cloud Agent Collaboration
Distributed architecture enabling privacy-preserving collaboration between enterprise and cloud AI agents.
3/10/2026
Disentangling Reasoning in Large Audio-Language Models for Ambiguous Emotion Prediction
Large audio-language models for ambiguous emotion recognition with disentangled reasoning.