Papers with Claude-3.5
Intrinsic Model Weaknesses: How Priming Attacks Unveil Vulnerabilities in Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Large language models (LLMs) have significant impact on various industries and societal functions due to advanced instruction-following capabilities. |
| Approach: | They developed and tested novel attack strategies on popular LLMs to expose their vulnerabilities in generating harmful content. |
| Outcome: | The proposed attacks achieved an ASR of 100% on open-source models, including Meta’s Llama-3.2, Google’s Gemma-2, Mistral’s Mistral-NeMo, Falcon’s Falcon-mamba, Apple’s DCLM, Microsoft’s Phi3, and Qwen’s Qwend2.5, among others. |
F2TEval: Human-Aligned Multi-Dimensional Evaluation for Figure-to-Text Task (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing evaluation methods for Figure-to-Text tasks are limited due to the inherent ambiguity and semantic compression of figures, the generated texts suffer from factual inaccuracies, incomplete coverage, and weak logical reasoning. |
| Approach: | They propose a five-dimensional reference-free evaluation method aligned with expert criteria to support fine-grained evaluation. |
| Outcome: | The proposed method outperforms Gemini-2.0 and Claude-3.5 with only 0.9B parameters. |
Automated Progressive Red Teaming (2025.coling-main)
Copied to clipboard
| Challenge: | Automated red teaming (ART) is effective but time-consuming, costly and lacks scalability. |
| Approach: | They propose an automated red teaming framework that generates adversarial prompts to expose LLM vulnerabilities. |
| Outcome: | The proposed framework explores and exploits LLM vulnerabilities through multi-round interactions. |
Navigating the Infinite Dynamic Web Space: Effective In-Context Exploration via Cognitive Multi-Agent Collaboration (2026.eacl-long)
Copied to clipboard
Guozhao Mo, Yanjiang Liu, Yafei Shi, Jiawei Chen, Yang Li, Yaojie Lu, Hongyu Lin, Ben He, Le Sun, Bo Zheng, Xianpei Han
| Challenge: | Existing methods for dynamic web navigation rely on greedy strategies or value estimation, struggle to achieve effective backtracking and are heavily dependent on proprietary models. |
| Approach: | They propose a cognitive multi-agent collaboration framework that enhances cyberspace exploration capability through In-Context Exploration. |
| Outcome: | The proposed framework surpasses the proprietary model Claude-3.5 Sonnet on the WebArena benchmark. |
Breaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to exploit black-box jailbreaks fail to capture key attack patterns . a novel framework decomposes jailbreak strategies into essential components . |
| Approach: | They propose a framework that decomposes jailbreak strategies into essential components and develops genetic-based optimization with intention evaluation mechanisms. |
| Outcome: | The proposed framework achieves 90% success rate on Claude-3.5, where prior methods completely fail . it also surpasses specialized safeguard models in evaluation accuracy . |
Language Models Can Infer Action Semantics for Symbolic Planners from Environment Feedback (2025.naacl-long)
Copied to clipboard
| Challenge: | Symbolic planners can discover sequences of actions from initial to goal states given domain-specific action semantics. |
| Approach: | They propose a domain induction task that requires an expert to define action semantics of domain to enable symbolic planning. |
| Outcome: | The proposed model boosts plan success rate from 36.4% to 100% on 7 environments. |
FC-Attack: Jailbreaking Multimodal Large Language Models via Auto-Generated Flowcharts (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Recent research shows that multimodal large language models are vulnerable to jailbreak attacks . |
| Approach: | They propose a jailbreak attack method based on auto-generated flowcharts . the flowchartings are then combined with a benign textual prompt to execute the attack . |
| Outcome: | The proposed method achieves an attack success rate of up to 96% via images and 78% via videos across multiple MLLMs. |
SemanticCamo: Jailbreaking Large Language Models through Semantic Camouflage (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have made safety issues of LLMs more prominent and critical. |
| Approach: | They propose a framework which attacks LLMs through semantic camouflage and replaces unsafe content with semantic features to conceal malicious intent . |
| Outcome: | The proposed framework outperforms existing models in over 80% of cases and is highly effective against various defenses. |
Can Multimodal Foundation Models Understand Schematic Diagrams? An Empirical Study on Information-Seeking QA over Scientific Papers (2025.findings-acl)
Copied to clipboard
| Challenge: | MISS-QA is the first benchmark specifically designed to evaluate the ability of models to interpret schematic diagrams within scientific literature. |
| Approach: | They propose an automated evaluation protocol powered by open-source LLMs trained on human-scored data to ensure reliable evaluation. |
| Outcome: | The proposed protocol is powered by open-source LLMs trained on human-scored data. |