arXiv:2606. 07555v1 Announce Type: cross Abstract: Glossaries, technical specifications, and system prompts routinely ask language models to use familiar words in unfamiliar ways.
arXiv:2606. 07555v5 Announce Type: replace-cross Abstract: Local definitions can assign a familiar word a temporary meaning while its usual associations remain useful elsewhere.
arXiv:2603. 00270v3 Announce Type: replace-cross Abstract: Large language models can process millions of tokens, yet how they handle conflicting information within context remains poorly understood.
arXiv:2608. 14681v1 Announce Type: cross Abstract: Words recur constantly in natural language use, yet it remains unclear whether language models reactivate prior representations or re-evaluate repeated words afresh, and whether post-training changes this default behavior.
arXiv:2608. 03297v1 Announce Type: new Abstract: A standard claim in the literature on retrieval-augmented and memory-augmented language models is that shorter context is better when the relevant information is preserved.
arXiv:2608. 04021v1 Announce Type: cross Abstract: Cloze-style probes that vary how often a target token appears implicitly assume that more copies of a target affect prediction the same way regardless of where the readout slot sits.
arXiv:2607. 13568v1 Announce Type: cross Abstract: Can a language model estimate its familiarity with an entity before generating an answer?
arXiv:2606. 12747v1 Announce Type: new Abstract: Safety-relevant studies of language models, including alignment and jailbreaking evaluations and AI control protocols, often rely on prefilling model outputs.
arXiv:2606. 03165v1 Announce Type: cross Abstract: The language used by digital chat assistants such as ChatGPT can diverge from human expectations (misalignment).
arXiv:2606. 08682v1 Announce Type: cross Abstract: Activation steering has emerged as a popular inference-time technique for modulating the behavior of large language models (LLMs).
arXiv:2607. 25907v1 Announce Type: cross Abstract: Activation steering controls model behavior by editing internal activations at inference time.
arXiv:2606. 11198v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) systems inject external knowledge to improve LLM outputs, yet the format of injected content -- distinct from its semantic relevance -- can independently distort the model's attention distribution.
arXiv:2607. 20524v1 Announce Type: new Abstract: Mean cross-positional attention degradation is widely reported in transformer interpretability, yet whether it causally limits contextual retrieval remains untested.