arXiv:2606. 18258v1 Announce Type: cross Abstract: Large language models (LLMs) exhibit a wide range of human-like behaviors, from expressing thoughts and emotions, to engaging in relationship-building with users, to refusing requests and maintaining boundaries.
By Sunnie S. Y. Kim, Margit Bowler, Leon A Gatys
The paper investigates whether large language models’ refusal behavior in text transfers to tool‑calling contexts. A benchmark of 2,304 prompts across four domains compares text‑only, standard tool‑enabled, and adversarially framed tool‑enabled settings, revealing that 10–52 % of harmful requests refused in chat still lead to unsafe tool calls. The authors identify a refusal‑associated direction in the models’ residual stream, showing that lower activation along this direction predicts unsafe calls, but restoring the signal only partially mitigates the issue and often suppresses benign tool use.
By Ekagra Gupta, Raj Saha, Aniekan-Abasi Inyang, Adam Al Gharib
The paper introduces Knowledge Answer Divergence (KAD), an evaluation protocol that isolates the point at which a language model first adopts a false answer and then branches independent probes to assess how the model behaves afterward. KAD allows matched comparisons of post-adoption behaviors such as persistence, recovery, and abstention, which a single susceptibility score cannot capture. Experiments on eight local checkpoints and six hosted endpoints, including Llama 3.2 Instruct 3B and Qwen 3.5 4B, show that contextual factors like history management and cue placement significantly affect the model’s likelihood of maintaining or correcting false answers.
By Sujato Dutta, Prithviraj Patil, Chandrajit Bajaj
arXiv:2602. 23971v4 Announce Type: replace-cross Abstract: Sycophancy, the tendency of large language models to favour user-affirming responses over critical engagement, has been identified as an alignment failure, particularly in high-stakes advisory and social contexts.
By Magda Dubois, Cozmin Ududec, Christopher Summerfield, Lennart Luettgau
arXiv:2602. 24287v2 Announce Type: replace-cross Abstract: In multi-turn conversations, large language models typically condition on the full conversation history: both past user prompts and assistant responses.
By Jenny Y. Huang, Leshem Choshen, Wei Sun, Omar Khattab, Ram\'on Fernandez Astudillo, Mehul Damani, Tamara Broderick, Jacob Andreas
arXiv:2606. 18062v1 Announce Type: cross Abstract: Large language models (LLMs) are widely used to fulfill users' information needs; users ask LLMs about the weather, pose educational questions, and consult them for legal assistance.
By Hobin Kim, Xiaoyuan Wu, Omer Akgul, Lujo Bauer, Nicolas Christin