arXiv AI
Jul 28
Tag Questions and the Generational Reversal of Sycophancy Across 45 Language Models
arXiv:2607. 23976v1 Announce Type: cross Abstract: Appending a two-word confirmation tag to a decision question -- "Is X the better choice?
Tag Questions and the Generational Reversal of Sycophancy Across 45 Language Models
Appending a two-word confirmation tag to a decision question -- "Is X the better choice? " versus "X is the better choice, right?
Related stories
arXiv Machine Learning
Jul 14
Two Confounds in Cross-Model Value Comparison: Response Determinism and the Access Harness
arXiv:2607. 10202v1 Announce Type: new Abstract: Cross-model comparisons read divergence in value dispositions as evidence that language models hold individuated values.
arXiv AI
Jul 15
The One-Word Census: Answer-Choice Conformity Across 44 Language Models
arXiv:2607. 12796v1 Announce Type: cross Abstract: When a language model must pick one answer from a large space of equally valid options, which does it pick -- and how often is it the same answer every other model picks?
Hugging Face Trending Papers
Jul 14
The One-Word Census: Answer-Choice Conformity Across 44 Language Models
When a language model must pick one answer from a large space of equally valid options, which does it pick -- and how often is it the same answer every other model picks? Asked to "pick a word -- any word," 44 models chose "serendipity" 41% of the time.
arXiv AI
Jul 21
Committed Before Reasoning: Behavioral Reproduction and Preliminary Activation-Level Evidence of Answer Pre-Commitment in an Open-Weight LLM
arXiv:2607. 16451v1 Announce Type: cross Abstract: Chat models sometimes commit to an answer and then produce reasoning that justifies it rather than deriving it -- even when the answer contradicts a task premise.
arXiv Machine Learning
Aug 11
When Counterbalancing Hides the Bias: Access-Conditioned Position Lock in Forced-Choice LLM Evaluation
arXiv:2607. 10202v2 Announce Type: replace Abstract: Forced-choice probes with counterbalanced orientations are a standard tool for measuring language-model "value dispositions," and a concentration/extremity index over repeated draws is read as how sharply a model commits.
arXiv AI
Jul 17
Breaking Refusal in the First Half: A Mechanistic Study of the Prefill Jailbreak
arXiv:2607. 14147v1 Announce Type: cross Abstract: Aligned language models refuse harmful requests, but a one-line prefill ("Sure, here is") strips the refusal.
arXiv AI
Jul 22
Structured Output Collapses Answer Diversity Across 44 Language Models
arXiv:2607. 18476v1 Announce Type: cross Abstract: When a language model must choose one answer from a large space of equally valid options, a format clause -- "Reply with JSON only" -- changes which answer it chooses.
arXiv AI
Jul 8
The yes-no bias of large language models reflects answer order and wording, not shifts in moral judgment
arXiv:2607. 05552v1 Announce Type: cross Abstract: Large language models (LLMs) increasingly issue judgments read as binary verdicts, and a growing literature reports such judgments shifting under logically irrelevant changes of wording - among them an amplified yes-no bias on moral dilemmas, absent in humans.
arXiv AI
Jul 29
Measuring and Improving Behavioral Consistency in Large Language Models through Fact-Heuristic-Emotion State Enforcement
arXiv:2607. 24765v1 Announce Type: cross Abstract: Large language models (LLMs) can give different answers to the same decision problem across runs, and reverse a decision when their own prior answer returns as context.
arXiv AI
Jul 7
Transplanting, inverting, and preventing a misalignment persona: method-conditional emergent misalignment in Qwen2.5
arXiv:2607. 04510v1 Announce Type: cross Abstract: Emergent misalignment (EM) -- the broad misbehaviour a language model acquires after fine-tuning on narrow harmful data -- is mediated in Qwen2.
arXiv Machine Learning
Jun 5
Measuring the sensitivity of LLM-based structured extraction to prompt, model, and schema choices in clinical discharge summaries
arXiv:2606. 05970v1 Announce Type: cross Abstract: Large language models are increasingly used for structured extraction from clinical free-text notes, but the sensitivity of their output to upstream configuration choices is less understood than their accuracy on fixed benchmarks.