Don’t Judge a Book by its Cover: Testing LLMs’ Robustness Under Logical Obfuscation (2026.eacl-long)
Copied to clipboard
| Challenge: | obfuscated questions pose significant challenges for large language models . current models parse questions without deep understanding, MIT researchers say . |
| Approach: | They propose a structure-preserving framework for logical obfuscation to test models . they use a logically equivalent framework to obliviate questions to logical equivalents . |
| Outcome: | The proposed framework is a first-of-its-kind diagnostic benchmark with 1,108 questions . obfuscation severely degrades zero-shot performance, the authors show . |
Similar Papers
Confusion is the Final Barrier: Rethinking Jailbreak Evaluation and Investigating the Real Misuse Threat of LLMs (2025.findings-emnlp)
Copied to clipboard
Yu Yan, Sheng Sun, Zhe Wang, Yijun Lin, Zenghao Duan, Zhifei Zheng, Min Liu, Zhiyi Yin, Jianping Zhang
| Challenge: | Large Language Models have been developed to deal with real-world crimes, but it remains unclear whether they internalize authentic knowledge or are forced to simulate toxic language patterns. |
| Approach: | They construct knowledge-intensive Q&A to investigate misuse threats of Large Language Models in terms of dangerous knowledge possession, harmful task planning utility, and harmfulness judgment robustness. |
| Outcome: | The findings raise concerns that jailbreak success is often attributable to a hallucination loop between jailbroken LLM and judger LLM . |
FOL-Traces: Verified First-Order Logic Reasoning Traces at Scale (2026.findings-eacl)
Copied to clipboard
| Challenge: | Existing approaches to evaluate language models fail to provide structural clarity and verifiable inference. |
| Approach: | They propose to use a large-scale dataset of programmatically verified reasoning traces to evaluate structured logical inference. |
| Outcome: | The proposed model achieves 45.7% accuracy on masked operation prediction and 27% on two-step completion. |
Safer or Luckier? LLMs as Safety Evaluators Are Not Robust to Artifacts (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly employed as automated evaluators to assess the safety of generated content. |
| Approach: | They evaluate 11 LLM judge models across critical safety domains . apologetic language artifacts alone can skew evaluator preferences by up to 98% . |
| Outcome: | The findings show that apologetic language artifacts alone can skew evaluator preferences by up to 98% . |
Root Defense Strategies: Ensuring Safety of LLM at the Decoding Level (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods to detect harmful outputs from prefill-level lacks utilization of the model’s decoding outputs, leading to relatively lower effectiveness and robustness. |
| Approach: | They propose a robust decoding mechanism that corrects harmful queries directly rather than rejecting them outright. |
| Outcome: | The proposed model improves model security without compromising reasoning speed. |
Truth or Sophistry? LoFa: A Benchmark for LLM Robustness Against Logical Fallacies (2026.acl-long)
Copied to clipboard
| Challenge: | Prior work has focused on the ability of Large Language Models to **identify** or **classify** fallacies, but their robustness against these fallacias in persuasive contexts remains largely unexplored. |
| Approach: | They propose a new metric to assess LLM robustness against fallacies by pairing factual questions with fallacious arguments and developing a multi-round debate framework to assess model resilience. |
| Outcome: | The proposed metric disentangles robustness from a model’s knowledge limitations and demonstrates unique vulnerability profiles across models. |
OBLIVIATE: Robust and Practical Machine Unlearning for Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) trained over corpora risk memorizing sensitive, copyrighted, or toxic content. |
| Approach: | They propose a framework that removes targeted data while preserving model utility. |
| Outcome: | The proposed framework resists membership inference attacks, minimizes impact on retained data, and maintains robustness across diverse scenarios. |
SecureSQL: Evaluating Data Leakage of Large Language Models as Natural Language Interfaces to Databases (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies on the vulnerability of large language models to SQL injection have been limited. |
| Approach: | They propose to evaluate the potential of language models to leak sensitive data when generating SQL queries. |
| Outcome: | The proposed model with the best performance has an accuracy of 61.7%, compared to humans who achieve 94% accuracy. |
Towards Data Contamination Detection for Modern Large Language Models: Limitations, Inconsistencies, and Oracle Challenges (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods for detecting data contamination in large language models have limitations and limitations . data contamination occurs when test or evaluation data is exposed to the model during its training phases . |
| Approach: | They evaluate five different methods for detecting data contamination in large language models . they find that current methods have non-trivial limitations in their assumptions and practical applications . |
| Outcome: | The proposed methods have non-trivial limitations and difficulties in detecting contamination . the authors highlight the complexity of contamination detection in advanced LLMs . |
CipherBank: Exploring the Boundary of LLM Reasoning Capabilities through Cryptography Challenge (2025.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) have demonstrated remarkable capabilities, but their capabilities in cryptographic decryption tasks remain underexplored. |
| Approach: | They propose a benchmark to evaluate the reasoning capabilities of large language models in cryptographic decryption tasks. |
| Outcome: | The proposed benchmark examines the reasoning capabilities of large language models in cryptographic decryption tasks. |
Diversity Helps Jailbreak Large Language Models (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods for jailbreaking large language models rely on laborious human engineering and whitebox access to model internals. |
| Approach: | They propose a method that instructs large language models to deviate from prior context and generate harmful outputs by instructing them to deviat from previous attacks. |
| Outcome: | The proposed method achieves a 62.83% higher success rate in compromising ten leading chatbots, while using only 12.9% of the queries. |