Sparse autoencoders inserted into VLMs and trained only for reconstruction can reliably detect adversarial attacks on images, including unseen domains and attack types.
Title resolution pending
16 Pith papers cite this work. Polarity classification is still indexing.
16
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
NeuralBench is a new benchmarking framework for neuroAI models on EEG data that finds foundation models only marginally outperform task-specific ones while many tasks like cognitive decoding stay highly challenging.
Image editing models fail zero-shot visual planning on abstract mazes and queen puzzles but generalize after finetuning, yet still cannot match human zero-shot efficiency.
PGT generates synthetic tasks via geometric overlays on images to supply dense visual supervision, improving spatial and relational understanding in MLLMs by up to 20% on targeted benchmarks.
PA-BDM adapts block diffusion by switching to causal intra-block denoising and dynamically committing reliable prefixes to KV cache, yielding higher accuracy and 71.6% higher throughput than a comparable baseline on document benchmarks.
Cumulative flow maps unify few-step generative modeling for diffusion and flow models via cumulative transport and parameterization with minimal changes to time embeddings and objectives.
Biased noise sampling for rectified flows combined with a bidirectional text-image transformer architecture yields state-of-the-art high-resolution text-to-image results that scale predictably with model size.
ALLaVA creates 1.3M GPT4V-synthesized samples enabling 4B VLMs to achieve competitive results on 17 benchmarks and match 7B/13B models on some tasks.
NeuralSet is a scalable Python framework that unifies diverse neural recordings and stimuli with deep learning embeddings via metadata decoupling and lazy data extraction.
A unified training framework for mesh-based ML surrogates in CFD improves accuracy and long-horizon stability by enforcing spatial derivative consistency via multi-node prediction, using temporal cross-attention correction, and adding 3D rotary positional embeddings.
VLMs recover reliable population-level trends in climate change visual discourse on social media even when per-image accuracy is only moderate.
citing papers explorer
-
Sparse Autoencoders as Plug-and-Play Firewalls for Adversarial Attack Detection in VLMs
Sparse autoencoders inserted into VLMs and trained only for reconstruction can reliably detect adversarial attacks on images, including unseen domains and attack types.
-
NeuralBench: A Unifying Framework to Benchmark NeuroAI Models
NeuralBench is a new benchmarking framework for neuroAI models on EEG data that finds foundation models only marginally outperform task-specific ones while many tasks like cognitive decoding stay highly challenging.
-
Probing Visual Planning in Image Editing Models
Image editing models fail zero-shot visual planning on abstract mazes and queen puzzles but generalize after finetuning, yet still cannot match human zero-shot efficiency.
-
PGT: Procedurally Generated Tasks for improving visual grounding in MLLMs
PGT generates synthetic tasks via geometric overlays on images to supply dense visual supervision, improving spatial and relational understanding in MLLMs by up to 20% on targeted benchmarks.
-
Prefix-Adaptive Block Diffusion for Efficient Document Recognition
PA-BDM adapts block diffusion by switching to causal intra-block denoising and dynamically committing reliable prefixes to KV cache, yielding higher accuracy and 71.6% higher throughput than a comparable baseline on document benchmarks.
-
A Few-Step Generative Model on Cumulative Flow Maps
Cumulative flow maps unify few-step generative modeling for diffusion and flow models via cumulative transport and parameterization with minimal changes to time embeddings and objectives.
-
Scaling Rectified Flow Transformers for High-Resolution Image Synthesis
Biased noise sampling for rectified flows combined with a bidirectional text-image transformer architecture yields state-of-the-art high-resolution text-to-image results that scale predictably with model size.
-
ALLaVA: Harnessing GPT4V-Synthesized Data for Lite Vision-Language Models
ALLaVA creates 1.3M GPT4V-synthesized samples enabling 4B VLMs to achieve competitive results on 17 benchmarks and match 7B/13B models on some tasks.
-
NeuralSet: A High-Performing Python Package for Neuro-AI
NeuralSet is a scalable Python framework that unifies diverse neural recordings and stimuli with deep learning embeddings via metadata decoupling and lazy data extraction.
-
Mesh Based Simulations with Spatial and Temporal awareness
A unified training framework for mesh-based ML surrogates in CFD improves accuracy and long-horizon stability by enforcing spatial derivative consistency via multi-node prediction, using temporal cross-attention correction, and adding 3D rotary positional embeddings.
-
From Codebooks to VLMs: Evaluating Automated Visual Discourse Analysis for Climate Change on Social Media
VLMs recover reliable population-level trends in climate change visual discourse on social media even when per-image accuracy is only moderate.
- Weasel: Out-of-Domain Generalization for Web Agents via Importance-Diversity Data Selection
- Unified Pix Token And Word Token Generative Language Model
- Online Learning-to-Defer with Varying Experts
- Reward Shaping and Action Masking for Compositional Tasks using Behavior Trees and LLMs
- Spherical Flows for Sampling Categorical Data