gyhdog99 / ECSO
ECSO (Make MLLM safe without neither training nor any external models!) (https://arxiv.org/abs/2403.09572)
☆9Updated 4 months ago
Related projects: ⓘ
- [ICLR 2024] Towards Elminating Hard Label Constraints in Gradient Inverision Attacks☆12Updated 7 months ago
- ☆24Updated 3 months ago
- ☆15Updated 4 months ago
- [ICML 2024] Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models.☆36Updated 3 weeks ago
- Instruction Tuning in Continual Learning paradigm☆19Updated 2 months ago
- [ECCV 2024] Official PyTorch Implementation of "How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs"☆61Updated 9 months ago
- Official code for ICLR 2024 paper, "A Hard-to-Beat Baseline for Training-free CLIP-based Adaptation"☆61Updated 4 months ago
- [ECCV 2024] The official code for "AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shi…☆34Updated 2 months ago
- [ICML 2024] Official implementation for "HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding"☆63Updated 4 months ago
- Accepted by ECCV 2024☆59Updated 2 months ago
- ☆35Updated last year
- Official PyTorch implementation of "CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning" @ ICCV 2023☆27Updated 8 months ago
- ☆32Updated last month
- Papers about Hallucination in Multi-Modal Large Language Models (MLLMs)☆43Updated 3 weeks ago
- [NeurIPS 2023] Generalized Logit Adjustment☆33Updated 4 months ago
- Official code for ICML 2024 paper, "Connecting the Dots: Collaborative Fine-tuning for Black-Box Vision-Language Models"☆13Updated 3 months ago
- HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction Data (Accepted by CVPR 2024)☆39Updated 2 months ago
- ☆16Updated last month
- Test-time Prompt Tuning (TPT) for zero-shot generalization in vision-language models (NeurIPS 2022))☆136Updated last year
- This is the official repo for Debiasing Large Visual Language Models, including a Post-Hoc debias method and Visual Debias Decoding strat…☆66Updated 5 months ago
- ☆24Updated last year
- ☆30Updated last month
- Official code for ICCV 2023 paper, "Improving Zero-Shot Generalization for CLIP with Synthesized Prompts"☆88Updated 6 months ago
- Exploring prompt tuning with pseudolabels for multiple modalities, learning settings, and training strategies.☆41Updated last week
- [ICLR 2024 Spotlight 🔥 ] - [ Best Paper Award SoCal NLP 2023 🏆] - Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal…☆14Updated 3 months ago
- IDEAL: Influence-Driven Selective Annotations Empower In-Context Learners in Large Language Models☆58Updated 8 months ago
- MMICL, a state-of-the-art VLM with the in context learning ability from ICL, PKU☆37Updated 11 months ago
- JailBreakV-28K: A comprehensive benchmark designed to evaluate the transferability of LLM jailbreak attacks to MLLMs, and further assess …☆29Updated 2 months ago
- HallE-Control: Controlling Object Hallucination in LMMs☆24Updated 5 months ago
- AnyDoor: Test-Time Backdoor Attacks on Multimodal Large Language Models☆39Updated 5 months ago