Scalable Weak-to-Strong Verifier Hardening for Agent Environments
Agent environments are essential for large language model (LLM) development. They use programmatic verifiers to turn complex outcomes into rewards. Verifiers can be attacked into a...
Research Papers
Latest AI/ML research from ArXiv and Papers with Code
CIViC: Evidence Gates for Audio-Conditioned Visual Prediction on Proxy Tasks
Removing audio can shift a visual predictor's output without showing whether the task uses the intended audio information. We introduce Conditional Input and Visual Interpretation ...
UniForge3D: Joint 3D Reconstruction and Generation in a Shared Anchor-Guided Space
Generating complete 3D assets from sparse, unposed images requires preserving observed geometry while completing unseen regions. Joint reconstruction and generation offers a way to...
ANTE: Anticipate, Then Execute -- Learning Internal World Models for Coding Agents from Verified Beliefs
Rapid advances in large language models have made software engineering agents capable of tackling complex coding tasks. Within a containerized environment, a coding agent receives ...
Agent Sociodrama: Social-Dynamics-Guided Persona Recasting for LLM Agent Teams
Large language model (LLM) agents are increasingly deployed as teams, yet they fail in characteristically social ways: information holders stay silent, dominant speakers anchor gro...
Apertus 1.5: From Open Text Models to Open Multimodal Reasoners
Language models are becoming infrastructure, and infrastructure built in the open, by public institutions, on documented data, is the alternative to dependence on a few providers; ...
Data-Centric Recursive Self-Improvement for Navigation via Policy-Synchronized Language-Action Dual Bootstrapping
Vision-language navigation requires agents to follow natural-language instructions to reach target locations, making high-quality language supervision essential. Since human annota...
SIMIT: Self-Improving Vision-Language Models via Imagination at Test-Time
We introduce SIMIT, a test-time self-improvement framework in which a vision-language model creates its own query-specific training data before answering an unlabeled test query. O...
A Decade in a Day: Synthesizing Agent-Native Interfaces from Human-Oriented Software Evolution
Large language model agents are emerging as an important new class of users of desktop applications. However, existing desktop interfaces were designed for human end users or appli...
VIEWCRAFT: Evolving What Multimodal Agents See and How They Read It
Reusable skills improve multimodal agents without updating model weights, but their effectiveness depends on both the evidence presented and the procedure used to read it. We intro...
Scaling Repeated-Data Pretraining: Batch Size, Regularization, and GPU-Time Tradeoffs
We study pretraining when compute is abundant but unique training data are finite, requiring repeated training on the same data. We first revisit infinite-compute pretraining (Kim ...
RSIPO: Recursive Self-Improving Policy Optimization for Agentic Reinforcement Learning via Skill-Policy Co-Evolution
Recent focus in agentic reinforcement learning (RL) seeks to improve credit assignment in long-horizon tasks. One promising direction combines reinforcement learning with verifiabl...
SkillMarker: A KGW-Based Watermarking Method for Agent Skill Ownership Verification
Agent skills encode operational procedures and domain expertise as reusable knowledge assets, yet their distribution in plaintext exposes them to unauthorized copying and leakage. ...
History Replays Itself: Unleashing Long-Context Capability in Linear Attention
Linear attention promises the efficiency needed for long-context language modelling, yet its fixed-size recurrent state must simultaneously support ongoing computation and preserve...
Exploring the Limits of Combinatorial Generalization in Embodied Manipulation
In embodied manipulation, Vision Language Action (VLA) models can execute atomic skills, but long-horizon manipulation requires recombining these skills without demonstrations for ...
Words to World: Benchmarking LLMs in Interactive Physical Environments
Words are increasingly becoming an interface for action: large language models (LLMs) can now invoke tools, operate devices, and make decisions that affect the physical world. Howe...
EviScene: Beyond "Is It AI?" for Agentic Image Factuality Assessment
Determining whether an image is AI-generated does not establish whether the situation it depicts is factually supported. We formulate Image Factuality Assessment: given an image an...
Drive3R: Block-wise Streaming Geometric Foundation Models for Autonomous Driving
High-fidelity closed-loop simulation for autonomous driving increasingly relies on dense multi-view 3D reconstruction. While recent feed-forward geometric foundation models have re...
Learning Where to Split: Dynamic Band Splitting for Audio Source Separation
Audio source separation models must preserve fine spectral structure while keeping the cost of the separator tractable. Band-split architectures reduce this cost by manually partit...
UltraBERT: Beyond the Limit of BERT Pre-Training
Recent bidirectional encoders have been modernized with innovations from causal large language models (LLMs). However, the pretraining recipe itself remains inherited from small-sc...
What Really Works for Universal Multimodal Embeddings? A Controlled Study of Contrastive Training Strategies
Contrastive learning has become the standard paradigm for multimodal representation learning, aligning semantically matched pairs while separating mismatched ones in a shared embed...
OmniCam: Omni-Camera Trajectory Generation via Geometry-Grounded Pose Token Learning
Camera trajectories control viewpoint changes in video generation, scene reconstruction, and robotic perception. Generating them from language requires both scene geometry and targ...
ProgAgent: Learning Tool Orchestration through Programmatic Training
Large language models (LLMs) are increasingly expected to orchestrate diverse tools over long-horizon tasks. Existing agents typically use Direct Tool Calling (DTC), where the mode...
LightVAE: Towards Compact and Efficient Video Autoencoders
Latent diffusion models enable high-quality video generation, but video VAE encoding and decoding remain computationally expensive. Existing methods use stage-specific lightweight ...