JAX-SPH: A Differentiable Smoothed Particle Hydrodynamics Framework
Comments Accepted at the ICLR 2024 Workshop on AI4Differential Equations In Science
期刊&会议
ICLR
International Conference on Learning Representations · 会议 · Machine Learning
Implicit regularization of deep residual networks towards neural ODEs
Comments ICLR 2024 (spotlight). 40 pages, 3 figures
Query-Policy Misalignment in Preference-Based Reinforcement Learning
Comments Accepted by ICLR 2024
Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting
Comments ICLR 2024 Camera Ready version. With respect to the original submission, we added text generation experiments, plots of entire accuracy distributions for each task + stdev computations, and prompt length correlation with spread analysis
On the Generalization and Approximation Capacities of Neural Controlled Differential Equations
Comments ICLR 2024. First presented at the F4CLD Workshop at ICML 2023
Increasing Model Capacity for Free: A Simple Strategy for Parameter Efficient Fine-tuning
Comments Accepted at ICLR 2024. Code at https://github.com/LINs-lab/CapaBoost
Does Writing with Language Models Reduce Content Diversity?
Comments ICLR 2024
KoLA: Carefully Benchmarking World Knowledge of Large Language Models
Comments Accepted by ICLR 2024
Distributed Inference Performance Optimization for LLMs on CPUs
Comments 4 pages, 3 figures, Practical ML for Low Resource Settings Workshop @ ICLR 2024
Multi-objective generative AI for designing novel brain-targeting small molecules
Comments 20 pages, 4 figures, Generative and Experimental Perspectives for Biomolecular Design Workshop at the 12th International Conference on Learning Representations
Consistent algorithms for multi-label classification with macro-at-$k$ metrics
Comments This is the authors' version of the work accepted to ICLR 2024; the final version of the paper, errors and typos corrected, and minor modifications to improve clarity
Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory
Comments 2024 ICLR Spotlight. The dataset and code can be found at https://confaide.github.io
Rethinking harmless refusals when fine-tuning foundation models
Comments ICLR 2024 AGI Workshop Poster
MIntRec2.0: A Large-scale Benchmark Dataset for Multimodal Intent Recognition and Out-of-scope Detection in Conversations
Comments Accepted by ICLR 2024, Long Paper; The abstract is slightly modified due to the length limitation
Mapping Land Naturalness from Sentinel-2 using Deep Contextual and Geographical Priors
Comments 6 pages, 3 figures, ICLR 2024 Tackling Climate Change with Machine Learning Workshop
Kernelised Normalising Flows
Comments Alternate title: Kernelized Normalizing Flows; Accepted at ICLR 2024
Lifting Architectural Constraints of Injective Flows
Comments Camera-ready version: accepted to ICLR 2024
Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task
Comments ICLR 2023 oral (notable-top-5%): https://openreview.net/forum?id=DeG07_TcZvT ; code: https://github.com/likenneth/othello_world
State Representation Learning Using an Unbalanced Atlas
Journal ref ICLR 2024
Shortened LLaMA: Depth Pruning for Large Language Models with Comparison of Retraining Methods
Comments Update (arXiv-v2): continued pretraining for severe pruning ratios, compatibility with quantization, and enhanced baselines. Preliminary work (arXiv-v1) accepted at ICLR 2024 Workshop on ME-FoMo: https://openreview.net/forum?id=18VGxuOdpu
Mixture of Weak & Strong Experts on Graphs
Comments Accepted for publication in ICLR 2024
Behaviour Distillation
Comments Published as a conference paper at ICLR 2024
Directly Fine-Tuning Diffusion Models on Differentiable Rewards
Comments Published at ICLR 2024
SRL: Scaling Distributed Reinforcement Learning to Over Ten Thousand Cores
Comments Published at ICLR 2024. 10 pages (24 pages with references and appendix), 7 figures
Evaluating representation learning on the protein structure universe
Comments ICLR 2024
A Primal-Dual Framework for Transformers and Neural Networks
Comments Accepted to ICLR 2023, 26 pages, 4 figures, 14 tables
Self-supervised Adversarial Training of Monocular Depth Estimation against Physical-World Attacks
Comments Accepted in TPAMI'24. Extended from our ICLR'23 publication (arXiv:2301.13487). arXiv admin note: substantial text overlap with arXiv:2301.13487
A Novel Sector-Based Algorithm for an Optimized Star-Galaxy Classification
Journal ref The Second Tiny Papers Track at ICLR 2024
Memorization in Self-Supervised Learning Improves Downstream Generalization
Comments Accepted at ICLR 2024
The Lie Derivative for Measuring Learned Equivariance
Comments ICLR 2023. Code available at: https://github.com/ngruver/lie-deriv
