Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks85 from public data
- 871
- Out-of-Distribution Detection and Selective Generation for Conditional Language Models241
This work presents a highly accurate and lightweight OOD detection method for CLMs, and shows its effectiveness on abstractive summarization and translation, and how it can be used under the common and realistic setting of distribution shift for selective generation of high-quality outputs, while automatically abstaining from low-quality ones, enabling safer deployment of generative language models.
- Plex: Towards Reliability using Pretrained Large Model Extensions150
Plex greatly improves the state-of-the-art across reliability tasks, and simplifies the traditional protocol as it improves the out-of-the-box performance and does not require designing scores or tuning the model for each task.
- Does your dermatology classifier know what it doesn’t know? Detecting the long-tail of unseen conditions124
A novel hierarchical outlier detection (HOD) loss is proposed, which assigns multiple abstention classes corresponding to each training outlier class and jointly performs a coarse classification of inliers vs. outliers, along with fine-grained classification of the individual classes.
- Retrieve-Rewrite-Answer: A KG-to-Text Enhanced LLMs Framework for Knowledge Graph Question Answering116
Experiments on several KGQA benchmarks show that the proposed KG-to-Text augmented LLMs approach outperforms previous KG -augmented LLMs approaches regarding answer accuracy and usefulness of knowledge statements.
- Self-Evaluation Improves Selective Generation in Large Language Models105
This work reformulates open-ended generation tasks into token-level prediction tasks, and leverage LLMs' superior calibration at the token level, and demonstrates that self-evaluation based scores not only improve accuracy, but also correlate better with the overall quality of generated content.
- DrugOOD: Out-of-Distribution Dataset Curator and Benchmark for AI-Aided Drug Discovery – a Focus on Affinity Prediction Problems with Noise Annotations81
DrugOOD offers an automated dataset curator with user-friendly customization scripts, rich domain annotations aligned with biochemistry knowledge, realistic noise level annotations, and rigorous benchmarking of SOTA OOD algorithms, as opposed to only providing fixed datasets.
- Improving Zero-shot Generalization and Robustness of Multi-Modal Models71
A simple and efficient zero-shot post-hoc method to identify images whose top-1 prediction is likely to be incorrect, by measuring consistency of the predictions w.r.t. multiple prompts and image transformations is developed.
- A Simple Zero-shot Prompt Weighting Technique to Improve Prompt Ensembling in Text-Image Models71
Using the proposed scoring method to create a weighted average prompt ensemble, the method outperforms equal average ensemble, as well as hand-crafted prompts, on ImageNet, 4 of its variants, and 11 fine-grained classification benchmarks, all while being fully automatic, optimization-free, and not requiring access to labeled validation data.
- 66
- DiffusionShield: A Watermark for Copyright Protection against Generative Diffusion Models65
A novel watermarking scheme, DiffusionShield, tailored for GDM s is introduced, which protects images from copyright infringement by GDM s through encoding the ownership information into an imperceptible watermark and injecting it into the images.
- Transferable Unlearnable Examples64
A novel unlearnable strategy based on Classwise Separability Discriminant (CSD) is proposed, which aims to better transfer the Unlearnable effects to other training settings and datasets by enhancing the linear separability.
- Explaining Neural Networks Semantically and Quantitatively61
This study proposes to distill knowledge from the CNN into an explainable additive model, which explains the CNN prediction quantitatively, and discusses the problem of the biased interpretation of CNN predictions.
- Defining and Quantifying the Emergence of Sparse Concepts in DNNs50
It is found that the inference score of a DNN can be disentangled into the effects of a few interactive concepts that can be understood as causal patterns in a sparse, symbolic causal graph, which explains the DNN.
- Towards an architectural treatment of software security42
The approach models security principal, privilege, trust, and context of architectural constituents, and can facilitate describing the security characteristics of an architecture generating enabling infrastructure, and monitoring run-time conformance.
- Learning to Remove: Towards Isotropic Pre-trained BERT Embedding41
In all tasks, the word embedding processed by the method consistently outperforms the original embedding (with average improvement of 13% on word analogy and 16% on semantic textual similarity) and two baseline methods and is proven to be more robust to changes of hyperparameter.
- 34
- FT-Shield: A Watermark Against Unauthorized Fine-tuning in Text-to-Image Diffusion Models31
The proposed FT-Shield addresses copyright protection challenges by designing new watermark generation and detection strategies and introduces an innovative algorithm for watermark generation that integrates a Mixture of Experts (MoE) approach for watermark detection.
- TorchOpt: An Efficient Library for Differentiable Optimization24
TorchOpt provides a unified and expressive differentiable optimization programming abstraction that allows users to efficiently declare and analyze various differentiable optimization programs with explicit gradients, implicit gradients, and zero-order gradients and provides a high-performance distributed execution runtime.
- 18
- Proving Common Mechanisms Shared by Twelve Methods of Boosting Adversarial Transferability17
This paper summarizes the common mechanism shared by twelve previous transferability-boosting methods in a unified view, i.e., these methods all reduce game-theoretic interactions between regional adversarial perturbations.
- Personalized Chinese Tourism Recommendation Algorithm Based on Knowledge Graph17
A new algorithm of personalized Chinese tourism recommendation based on the Knowledge Graph with significant improvement over the existing algorithms is proposed and the method of user interest modelbased on the attribute information of users and tourist attractions is proposed to improve the performance of the recommendation system.
- 15
- 14
- Mining Interpretable AOG Representations From Convolutional Networks via Active Question Answering14
A question-answering method that uses active human-computer communications to mine patterns from a pre-trained CNN, in order to incrementally explain more features in conv-layers, and achieves similar or better part-localization performance than fast-RCNN methods.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.