Deep Learning
OpenAI is hiring a new Head of Preparedness to try to predict and mitigate AI's harms
Switch 2 games are on sale through Jan. 5 OpenAI is hiring a new Head of Preparedness to try to predict and mitigate AI's harms CEO Sam Altman posted about the role on X, saying the models'are starting to present some real challenges.' OpenAI is looking for a new Head of Preparedness who can help it anticipate the potential harms of its models and how they can be abused, in order to guide the company's safety strategy. It comes at the end of a year that's seen OpenAI hit with numerous accusations about ChatGPT's impacts on users' mental health, including a few wrongful death lawsuits . In a post on X about the position, OpenAI CEO Sam Altman acknowledged that the potential impact of models on mental health was something we saw a preview of in 2025, along with other real challenges that have arisen alongside models' capabilities. The Head of Preparedness is a critical role at an important time, he said.
Passive learning of active causal strategies in agents and language models
What can be learned about causality and experimentation from passive data?
Training Your Image Restoration Network Better with Random Weight Network as Optimization Function
The blooming progress made in deep learning-based image restoration has been largely attributed to the availability of high-quality, large-scale datasets and advanced network structures. However, optimization functions such as L 2 are still de facto. In this study, we propose to investigate new optimization functions to improve image restoration performance. Our key insight is that ``random weight network can be acted as a constraint for training better image restoration networks''. However, not all random weight networks are suitable as constraints.
An End-to-End Graph Attention Network Hashing for Cross-Modal Retrieval
Due to its low storage cost and fast search speed, cross-modal retrieval based on hashing has attracted widespread attention and is widely used in real-world applications of social media search.
DiffusionPID: Interpreting Diffusion via Partial Information Decomposition
Text-to-image diffusion models have made significant progress in generating naturalistic images from textual inputs, and demonstrate the capacity to learn and represent complex visual-semantic relationships. While these diffusion models have achieved remarkable success, the underlying mechanisms driving their performance are not yet fully accounted for, with many unanswered questions surrounding what they learn, how they represent visual-semantic relationships, and why they sometimes fail to generalize.
Structured flexibility in recurrent neural networks via neuromodulation
Such models range broadly in both complexity and biological plausibility.
AdaNovo: Towards Robust De Novo Peptide Sequencing in Proteomics against Data Biases Jun Xia
Despite the development of several deep learning methods for predicting amino acid sequences (peptides) responsible for generating the observed mass spectra, training data biases hinder further advancements of de novo peptide sequencing. Firstly, prior methods struggle to identify amino acids with Post-Translational Modifications (PTMs) due to their lower frequency in training data compared to canonical amino acids, further resulting in unsatisfactory peptide sequencing performance. Secondly, various noise and missing peaks in mass spectra reduce the reliability of training data (Peptide-Spectrum Matches, PSMs).