sail-sg / Cheating-LLM-Benchmarks
[ICLR 2025] Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates (Oral)
☆75Updated 5 months ago
Alternatives and similar repositories for Cheating-LLM-Benchmarks:
Users that are interested in Cheating-LLM-Benchmarks are comparing it to the libraries listed below
- Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses (NeurIPS 2024)☆59Updated 2 months ago
- [ICLR'24] RAIN: Your Language Models Can Align Themselves without Finetuning☆89Updated 10 months ago
- An official implementation of "Catastrophic Failure of LLM Unlearning via Quantization" (ICLR 2025)☆26Updated last month
- Code for safety test in "Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates"☆18Updated last year
- Weak-to-Strong Jailbreaking on Large Language Models☆72Updated last year
- Official Repository for The Paper: Safety Alignment Should Be Made More Than Just a Few Tokens Deep☆83Updated 8 months ago
- Codebase for decoding compressed trust.☆23Updated 10 months ago
- [ICLR 2025] When Attention Sink Emerges in Language Models: An Empirical View (Spotlight)☆54Updated 5 months ago
- Implementation of PaCE: Parsimonious Concept Engineering for Large Language Models (NeurIPS 2024)☆35Updated 4 months ago
- ☆31Updated last year
- Code for "Reasoning to Learn from Latent Thoughts"☆77Updated this week
- [ICLR 2025] Official Repository for "Tamper-Resistant Safeguards for Open-Weight LLMs"☆49Updated last month
- ☆53Updated 2 years ago
- Safe Unlearning: A Surprisingly Effective and Generalizable Solution to Defend Against Jailbreak Attacks☆25Updated 8 months ago
- ☆46Updated last year
- [ICML 2024] Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications☆73Updated last month
- ☆31Updated 2 months ago
- ☆37Updated last year
- ☆42Updated last month
- Official implementation of Bootstrapping Language Models via DPO Implicit Rewards☆43Updated 8 months ago
- ☆26Updated 9 months ago
- ☆34Updated 6 months ago
- ☆82Updated 2 weeks ago
- Repo for the research paper "SecAlign: Defending Against Prompt Injection with Preference Optimization"☆41Updated 2 months ago
- The official implementation of "LightTransfer: Your Long-Context LLM is Secretly a Hybrid Model with Effortless Adaptation"☆14Updated 2 weeks ago
- Röttger et al. (NAACL 2024): "XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models"☆95Updated last month
- Stanford NLP Python library for benchmarking the utility of LLM interpretability methods☆62Updated this week
- ConceptVectors Benchmark and Code for the paper "Intrinsic Evaluation of Unlearning Using Parametric Knowledge Traces"☆33Updated last month
- The official repository of 'Unnatural Language Are Not Bugs but Features for LLMs'☆13Updated 3 weeks ago
- Package to optimize Adversarial Attacks against (Large) Language Models with Varied Objectives☆67Updated last year