AI &amp; ML Research Papers

Controllable video generation: A survey

Unknown

A systematic review of controllable video generation, covering methods, open-source models, and future directions.

0Jul 1, 2025

Preprint2024

Controlvideo: Training-free controllable text-to-video generation

Unknown

Controlvideo introduces a training-free framework for controllable text-to-video generation by leveraging pre-trained text-to-image models.

0Jan 1, 2024

Reflections on the AI alignment problem

Unknown

This paper reflects on the AI alignment problem, discussing the challenges of ensuring AGI conforms to human values and remains under human control.

0Jan 1, 2025Alignment

Detecting benchmark contamination through watermarking

Unknown

This paper proposes a watermarking-based method to detect benchmark contamination in large language models, validated on 10B tokens with controlled contamination.

0Feb 1, 2025Benchmarks

0Aug 1, 2025Few Shot LearningComputer Vision

Few-shot neuro-symbolic imitation learning for long-horizon planning and acting

Unknown

This paper introduces a few-shot neuro-symbolic imitation learning framework that combines symbolic planning with neural control for long-horizon tasks.

0Jan 1, 2026Large Language Models

Decoding-Time Debiasing via Process Reward Models: From Controlled Fill-in to Open-Ended Generation

Unknown

This paper introduces a decoding-time debiasing method using process reward models to reduce social biases in LLM generation, from fill-in tasks to open-ended generation.

PreprintarXiv.org2025

Reasoning on a budget: A survey of adaptive and controllable test-time compute in llms

Mohammad Ali Alomrani, Yingxue Zhang, Derek Li, et al.

A survey of adaptive and controllable test-time compute methods in LLMs, focusing on efficiency across different compute paradigms.

25Jul 1, 2025Reasoning

Molmoact: Action reasoning models that can reason in space

Unknown

Introduces Action Reasoning Models (ARMs), a robotic foundation model class with a structured three-stage pipeline integrating perception, planning, and control.

0Aug 1, 2025Reasoning

PreprintMaastricht Journal of European and Comparative Law2020

Proceduralizing control and discretion: Human oversight in artificial intelligence policy

Riikka Koulu

This paper critically examines human oversight in EU AI policy for legal decision-making, arguing it risks becoming an empty procedural safeguard without addressing inherent human limitations.

43Dec 1, 2020

0Jul 30, 2026Ai SafetyBenchmarks

PAC-MAN: Perception-Aware CBF-RL for Whole-Body Safety in Humanoid Dodgeball

Lizhi Yang, Junheng Li, Aaron D. Ames

PAC-MAN couples control-barrier safety with perception-aware training for whole-body humanoid dodgeball, achieving 95% real-world evasion with a fixed camera.

Beyond Geometric Complementarity: Coherent Overlap in Sparse Mixture-of-Experts Routing

Huiyuan Tian, Bonan Xu, Shijian Li

This paper introduces the Expert Subspace Separation Index (ESSI) and controlled experiments to show that sparse MoE routing selects token-relevant experts from overlapping subspaces, a pattern termed 'coherent overlap' that preserves functional valu

0Jul 30, 2026

0Jul 30, 2026Reinforcement LearningReasoning

β-OPSD: Deriving with Policy Optimization, Training with Self-Distillation

Jiawei Xu, Minghui Liu, Juzheng Zhang, et al.

β-OPSD generalizes on-policy self-distillation by introducing a controllable KL penalty weight β, enabling efficient distillation to approximate policy optimization and improving reasoning performance.