Frontier Atlas - Discover AI Research

OpenMedQ: Broad Open Pretraining for Medical Vision-Language Models

Max Van Puyvelde, Olivier Gevaert, Ibrahim Gulluk·Jun 11, 2026

Foundation Models

Chain of Thought

Exploding and vanishing gradients in deep neural networks: the effect of residual connections

Vivek S Borkar·Jun 15, 2026

Back to the Familiar Future: Failure Recovery for VLA Policies via Pre-Imagined Milestone Selection

Hyung-Sin Kim, Suyeon Shin, Hyundo Lee et al.·Jun 8, 2026

Rethinking LoRA Memory Through the Lens of KV Cache Compression

Benjamin Van Durme, Liaoyaqi Wang, William Jurayj et al.·Jun 4, 2026

Foundation Models

Diffusion Transformer

The Hidden Power of Scaling Factor in LoRA Optimization

Haoran Li, Zicheng Zhang, Anqi Li et al.·Jun 11, 2026

Beyond Uniform Forgetting: A Study of Sequential Direct Preference Optimization Across Preference Settings

Pranav Bhandari, Mehwish Nasim, Nicolas Fay et al.·Jun 18, 2026

Exploring Extrinsic and Intrinsic Properties for Effective Reasoning with Code Interpreter

Pittawat Taveekitworachai, Panjapong Poobanchuen, Samuel Cahyawijaya et al.·Jun 15, 2026

WISE: A Long-Horizon Agent in Minecraft with Why-Which Reasoning

Renmin Cheng, Changhao Chen·Jun 11, 2026

SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents

Hanting Chen, Xinghao Chen, Haocheng Mei et al.·Jun 16, 2026

Are Text-to-Image Models Inductivist Turkeys? A Counterfactual Benchmark for Causal Reasoning

Rongpeng Zhu, Zijian Zhou, Bin Fu et al.·Jun 23, 2026

RetailBench: Benchmarking long horizon reasoning and coherent decision making of LLM agents in realistic retail environments

Jingtong Wu, Linghua Zhang, Jun Wang et al.·Jun 14, 2026

Foundation Models

3D Gaussian Splatting

QPredSGG: Hybrid Quantum Predicate Learning for Long-Tailed Scene Graph Generation

Prerana Ramkumar, Nouhaila Innan, Muhammad Shafique·Jun 3, 2026

Graph Grounded Cross Attention Transformer Neural Network for Structurally Constrained Full Event Sequence Generation in Predictive Process Monitoring

Fang Wang, Ernesto Damiani·Jun 17, 2026

PAI: Preserving Amplitude Information in Representation-Based Time-Series Anomaly Detection

Shoushou Ren, Dong Lin, Joon Son Chung et al.·Jun 8, 2026

NeRD: Neuro-Symbolic Rule Distillation for Efficient Ontology-Grounded Chain-of-Thought in Medical Image Diagnosis

Siyuan Yan, Eunis Li, Zongyuan Ge et al.·Jun 14, 2026

Foundation Models

Quantum Neural Network

How Requirements Quality Makes (or Breaks) Traceability Link Recovery

Julian Frattini, Tobias Hey·Jun 10, 2026

From Uncertain Judgments to Calibrated Rankings: Conformal Elo Estimation for LLM Evaluation

David Salinas, Bora Kargi·Jun 11, 2026

TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning

Lizhou Cai, Yuhang Jiang, Weijie Liu et al.·Jun 9, 2026

Foundation Models

Quantum Neural Network

Leverage Is Not Reach: A Control-Window Law for Single-Neuron Steering in Language Models

Hongliang Liu·Jun 18, 2026

Thinking with Visual Grounding

Wei Wang, Junkai Zhang, Kai-Wei Chang et al.·Jun 15, 2026

FOUNDATION MODELS

OpenMedQ: Broad Open Pretraining for Medical Vision-Language Models

Exploding and vanishing gradients in deep neural networks: the effect of residual connections

Back to the Familiar Future: Failure Recovery for VLA Policies via Pre-Imagined Milestone Selection

Rethinking LoRA Memory Through the Lens of KV Cache Compression

The Hidden Power of Scaling Factor in LoRA Optimization

Beyond Uniform Forgetting: A Study of Sequential Direct Preference Optimization Across Preference Settings

Exploring Extrinsic and Intrinsic Properties for Effective Reasoning with Code Interpreter

WISE: A Long-Horizon Agent in Minecraft with Why-Which Reasoning

SafeClawBench: Separating Semantic, Audit-Evidence, and Sandbox Harm in Tool-Using LLM Agents

Are Text-to-Image Models Inductivist Turkeys? A Counterfactual Benchmark for Causal Reasoning

RetailBench: Benchmarking long horizon reasoning and coherent decision making of LLM agents in realistic retail environments

QPredSGG: Hybrid Quantum Predicate Learning for Long-Tailed Scene Graph Generation

Graph Grounded Cross Attention Transformer Neural Network for Structurally Constrained Full Event Sequence Generation in Predictive Process Monitoring

PAI: Preserving Amplitude Information in Representation-Based Time-Series Anomaly Detection

NeRD: Neuro-Symbolic Rule Distillation for Efficient Ontology-Grounded Chain-of-Thought in Medical Image Diagnosis

How Requirements Quality Makes (or Breaks) Traceability Link Recovery

From Uncertain Judgments to Calibrated Rankings: Conformal Elo Estimation for LLM Evaluation

TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning

Leverage Is Not Reach: A Control-Window Law for Single-Neuron Steering in Language Models

Thinking with Visual Grounding