Greedy Coordinate Gradient
PulseAugur coverage of Greedy Coordinate Gradient — every cluster mentioning Greedy Coordinate Gradient across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New Reflex-Guard system offers low-latency LLM prompt safety
Researchers have developed Reflex-Guard, a novel local guardrail system designed to enhance the safety of Large Language Models (LLMs) without introducing significant latency or privacy concerns. Unlike existing methods…
-
New C2A model improves chest X-ray classification by coupling spatial and clinical data
Researchers have developed a new classification head called C$^2$A (Co-occurrence Aware Class Attention) designed to improve the accuracy of multi-label classification for chest X-rays. This method explicitly links spat…
-
New framework offers realistic safety guarantees for LLMs
Researchers have developed a new probabilistic framework, termed "(k, \epsilon)-unstable," to provide more realistic safety guarantees for Large Language Models (LLMs) against jailbreaking attacks. This approach improve…
-
Open-source LLM agents vulnerable to financial loss via adversarial attacks
LLM agents that manage significant financial assets are vulnerable to gradient-based adversarial attacks due to their reliance on open-source models like LLaMA and Mistral AI. Attackers can download model weights and cr…
-
New "Sockpuppetting" Attack Method Exploits LLM Vulnerabilities
Researchers have developed a new method called "sockpuppetting" to bypass safety measures in large language models. This technique combines prefill attacks, which insert an acceptance sequence at the beginning of an LLM…
-
New research probes and breaks LLM safety mechanisms
Researchers have developed new methods to probe and potentially break the safety mechanisms in large language models. By using activation-guided adversarial suffixes, they found that safety representations are distribut…
-
New methods probe and break LLM safety representations
Researchers have developed new methods to probe and potentially break the safety mechanisms within large language models. By analyzing how models refuse certain prompts, they identified that safety representations are d…
-
New STEER attack exploits LLM safety gaps in multilingual contexts · 3 sources tracked
Researchers have developed a new method called STEER (Safety Targeted Embedding Exploit via Refinement) to exploit vulnerabilities in the safety training of large language models (LLMs). This technique targets models tr…
-
New research explores multimodal and sparse autoencoder methods to combat LLM jailbreaks
Researchers are developing new methods to combat jailbreaking attacks on spoken language models (SLMs). One approach, JAMA, uses a joint multimodal optimization framework to simultaneously attack both audio and text mod…
-
New SCARCE method improves rare-event analysis in AI systems
Researchers have introduced SCARCE (Scalable Cascade Analysis for Rare-event Characterisation via Embeddings), a novel method for estimating the probabilities of rare events in AI systems. SCARCE replaces traditional pe…
-
New response-time probing method boosts LLM safety against prefilling attacks
Researchers have developed a new method called response-time probing to enhance the safety of large language models by detecting prefilling attacks. This technique, which probes the model's hidden state at the first gen…
-
New ASR techniques tackle phonetic errors and judge reliability
Researchers are developing advanced methods to improve Automatic Speech Recognition (ASR) systems, particularly for low-resource languages and to address specific types of errors. One approach, Error-Aware TF-IDF, uses …
-
New defenses and attacks target LLM jailbreaks and prompt injections
Researchers are developing new methods to defend large language models against prompt injection and jailbreak attacks. GuardNet utilizes an ensemble of shallow neural networks for efficient detection, while SlotGCG focu…
-
New research reveals escalating LLM and LALM jailbreak vulnerabilities
Three new research papers explore the vulnerabilities and defenses of large language models (LLMs) and large audio-language models (LALMs). The first paper details a taxonomy of audio jailbreak attacks and defenses, hig…
-
New Frost Training method boosts LLM policy optimization
Researchers have introduced Frost Training, a novel method designed to enhance Monte Carlo-based policy optimization for a class of tasks known as Cross-Entropy Games. This technique leverages the gradient of the reward…
-
New Logit-Gap Steering method efficiently measures AI alignment robustness
Researchers have developed a new metric called the refusal-affirmation logit gap to quantify the safety margin of aligned language models. This metric, which measures the difference between refusal and affirmation token…
-
Researchers explore token position's impact on LLM adversarial attacks
Researchers have identified a critical blind spot in the adversarial robustness evaluation of large language models. Their study, focusing on the Greedy Coordinate Gradient (GCG) attack, reveals that the placement of ad…