Grammar-Constrained Decoding Can Jailbreak LLMs into Generating Malicious Code
Paper • 2606.11817 • Published • 17 •
3
Noah
noahml
AI & ML interests
None yet
Recent Activity
commentedon a paper about 16 hours ago
Grammar-Constrained Decoding Can Jailbreak LLMs into Generating Malicious Code commentedon a paper about 16 hours ago
Reroute, Don't Remove: Recoverable Visual Token Routing for Vision-Language Models commentedon a paper about 16 hours ago
TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement LearningOrganizations
None yet
commented 5
papers about 16 hours ago
Reroute, Don't Remove: Recoverable Visual Token Routing for Vision-Language Models
Paper • 2606.12412 • Published • 15 •
3
TRACE: A Unified Rollout Budget Allocation Framework for Efficient Agentic Reinforcement Learning
Paper • 2606.11119 • Published • 17 •
3
Robust-U1: Can MLLMs Self-Recover Corrupted Visual Content for Robust Understanding?
Paper • 2606.08063 • Published • 71 •
3
VIA-SD: Verification via Intra-Model Routing for Speculative Decoding
Paper • 2606.12243 • Published • 14 •
3
commented 7
papers 1 day ago
WorldOlympiad: Can Your World Model Survive a Triathlon?
Paper • 2606.11129 • Published • 29 •
3
In-Context Multiple Instance Learning
Paper • 2606.06458 • Published •
3
PsychoSafe: Eliciting Psychologically-Informed Refusals in Large Language Models
Paper • 2606.09697 • Published • 6 •
4
UniPET: a universal network for high-quality PET image denoising across varied dose reduction factors
Paper • 2606.11131 • Published • 5 •
3
Redesign Mixture-of-Experts Routers with Manifold Power Iteration
Paper • 2606.12397 • Published • 81 •
4
Toward Generalist Autonomous Research via Hypothesis-Tree Refinement
Paper • 2606.11926 • Published • 104 •
6
On Subquadratic Architectures: From Applications to Principles
Paper • 2606.12364 • Published • 22 •
3
commented 7
papers 3 days ago
OmniCap-IF: Benchmarking and Improving Instruction Following Abilities for Omni-Video Captioning
Paper • 2606.08572 • Published • 14 •
3
A Geometric Account of Activation Steering through Angle-Norm Decomposition
Paper • 2606.06735 • Published • 21 •
3
Whisper Hallucination Detection and Mitigation via Hidden Representation Steering and Sparse AutoEncoders
Paper • 2606.07473 • Published • 15 •
3
Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories
Paper • 2606.11176 • Published • 41 •
2
SearchSwarm: Towards Delegation Intelligence in Agentic LLMs for Long-Horizon Deep Research
Paper • 2606.09730 • Published • 49 •
3
Direct 3D-Aware Object Insertion via Decomposed Visual Proxies
Paper • 2606.06601 • Published • 26 •
5
UnpredictaBench: A Benchmark for Evaluating Distributional Randomness in LLMs
Paper • 2606.06622 • Published • 19 •
3
commented a
paper 4 days ago