Apple

company

Verified

https://www.apple.com/

AI & ML interests

None defined yet.

Recent Activity

rezaarmand submitted a paper about 3 hours ago

Unmasking On-Policy Distillation: Where It Helps, Where It Hurts, and Why

seyedhamidreza submitted a paper about 10 hours ago

A Single Neuron Is Sufficient to Bypass Safety Alignment in Large Language Models

shaokaiyeah submitted a paper 1 day ago

BalCapRL: A Balanced Framework for RL-Based MLLM Image Captioning

View all activity

Papers

Unmasking On-Policy Distillation: Where It Helps, Where It Hurts, and Why

A Single Neuron Is Sufficient to Bypass Safety Alignment in Large Language Models

View all Papers

Team members 792
private

apple 's Papers 38

Submitted by

Mohammadreza Armandpour

Unmasking On-Policy Distillation: Where It Helps, Where It Hurts, and Why

apple

Submitted by

Hamid Kazemi

A Single Neuron Is Sufficient to Bypass Safety Alignment in Large Language Models

apple

Submitted by

Amin Karimi Monsefi

Trajectory as the Teacher: Few-Step Discrete Flow Matching via Energy-Navigated Distillation

apple

Submitted by

Shaokai Ye

BalCapRL: A Balanced Framework for RL-Based MLLM Image Captioning

apple

2

Submitted by

taesiri

STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation

apple

Submitted by

taesiri

Normalizing Trajectory Models

apple

Submitted by

Ajay Jaiswal

TIDE: Every Layer Knows the Token Beneath the Context

apple

Submitted by

taesiri

Stochastic KV Routing: Enabling Adaptive Depth-Wise Cache Sharing

apple

Submitted by

taesiri

Embarrassingly Simple Self-Distillation Improves Code Generation

apple

Submitted by

taesiri

Less Gaussians, Texture More: 4K Feed-Forward Textured Splatting

apple

Submitted by

Tzu-Heng Huang

RubiCap: Rubric-Guided Reinforcement Learning for Dense Image Captioning

apple

Submitted by

Parshin Shojaee

Recursive Language Models Meet Uncertainty: The Surprising Effectiveness of Self-Reflective Program Search for Long Context

apple

2

Submitted by

Xinyu Zhu

CHIMERA: Compact Synthetic Data for Generalizable LLM Reasoning

apple

3

Submitted by

taesiri

The Design Space of Tri-Modal Masked Diffusion Models

apple

Submitted by

taesiri

On Robustness and Chain-of-Thought Consistency of RL-Finetuned VLMs

apple

Submitted by

Ajay Jaiswal

MemoryLLM: Plug-n-Play Interpretable Feed-Forward Memory for Transformers

apple

Submitted by

taesiri

GenCtrl -- A Formal Controllability Toolkit for Generative Models

apple

Submitted by

taesiri

Over-Searching in Search-Augmented Large Language Models

apple

Submitted by

Amael Delaunoy

Sharp Monocular View Synthesis in Less Than a Second

apple

Submitted by

taesiri

Learning Unmasking Policies for Diffusion Language Models

apple

Submitted by

Jiatao Gu

One Layer Is Enough: Adapting Pretrained Visual Encoders for Image Generation

apple

2

Submitted by

Jiatao Gu

STARFlow-V: End-to-End Video Generative Modeling with Normalizing Flow

apple

Submitted by

jie he

CLaRa: Bridging Retrieval and Generation with Continuous Latent Reasoning

apple

Submitted by

taesiri

SO-Bench: A Structural Output Evaluation of Multimodal LLMs

apple

Submitted by

Feijie Wu

PORTool: Tool-Use LLM Training with Rewarded Tree

apple

1

Submitted by

taesiri

PRISM-Bench: A Benchmark of Puzzle-Based Visual Tasks with CoT Error Detection

apple

Submitted by

taesiri

Pico-Banana-400K: A Large-Scale Dataset for Text-Guided Image Editing

apple

Submitted by

taesiri

UltraCUA: A Foundation Model for Computer Use Agents with Hybrid Action

apple

Submitted by

Nikhil

Mirror Speculative Decoding: Breaking the Serial Barrier in LLM Inference

apple

Submitted by

taesiri

DeepMMSearch-R1: Empowering Multimodal LLMs in Multimodal Web Search

apple

Submitted by

Guan

Taming Text-to-Sounding Video Generation via Advanced Modality Condition and Interaction

apple

2

Submitted by

Huangjie Zheng

Continuously Augmented Discrete Diffusion model for Categorical Generative Modeling

apple

Submitted by

taesiri

Ferret-UI Lite: Lessons from Building Small On-Device GUI Agents

apple

Submitted by

Shenao Zhang

Learning to Reason as Action Abstractions with Scalable Mid-Training RL

apple

2

Submitted by

Hadi Pouransari

Pretraining with hierarchical memories: separating long-tail and common knowledge

apple

Submitted by

xhl

Rethinking JEPA: Compute-Efficient Video SSL with Frozen Teachers

apple

2

LinEAS: End-to-end Learning of Activation Steering with a Distributional Loss

apple

Submitted by

Pau Rodriguez

Controlling Language and Diffusion Models by Transporting Activations

apple