arXiv AI
Jul 29
Similar Models Learn Differently: Final-Window Pretraining Shapes Post-Training Beyond SFT
arXiv:2607. 25063v1 Announce Type: new Abstract: Developers judge a model checkpoint by how it behaves.
Refusal Reads Only a Slice of What the Model Knows: Harm-Keyed Routing and Its Exceptions Across Model Families
Related stories
arXiv AI
Sep 15
Calibrating Interpretability Instruments Before Trusting Their Verdicts
arXiv:2609.14754v1 Announce Type: cross Abstract: Causal claims about large language model (LLM) internals rest on measurements. Those might include a projection, a cosine, an ablation delta, or an i...
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 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.
Hugging Face Trending Papers
Jul 6
The yes-no bias of large language models reflects answer order and wording, not shifts in moral judgment
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. A single framing cannot say what such a shift is: in a yes/no question the word "no" is at once logical verdict, lexical token, and last-printed option.
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 AI
3d ago
Unlearning Deceptive Behaviors in LLMs with Contrastive Forget Sets
The paper introduces PACT, a method for unlearning deceptive behaviors in large language models by using contrastive forget sets that compare a model’s responses under deceptive and neutral contexts. PACT trains the model to produce pressure‑aware counterfactual targets, preserving benign system‑prompt adherence and reasoning traces while dramatically reducing deception rates from over 50% to under 3% on 32B reasoning models.
arXiv Machine Learning
Jul 20
Beyond a Single Direction: Chain-of-Thought Disrupts Simple Steering of Refusal
arXiv:2605. 26772v1 Announce Type: cross Abstract: Large reasoning models (LRMs) generate chain-of-thought (CoT) traces before producing final outputs, introducing a dynamic internal state that may complicate control mechanisms such as refusal.
arXiv Machine Learning
Jun 5
Moral Sensitivity in LLMs: A Tiered Evaluation of Contextual Bias via Behavioral Profiling and Mechanistic Interpretability
arXiv:2605. 03217v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly deployed in settings that require nuanced ethical reasoning, yet existing bias evaluations treat model outputs as simply "biased" or "unbiased.
arXiv AI
Jul 22
Measuring Reward-Seeking via Contrastive Belief Updates
arXiv:2607. 18966v1 Announce Type: new Abstract: Language models trained with reinforcement learning may learn to optimize the grader's judgment rather than the intended objective.
arXiv Machine Learning
Sep 22
Knowing, and Saying It Only When Asked: LLM Endognostics and the Schizognosis of Minerva-7B
arXiv:2609.22219v1 Announce Type: cross Abstract: Evaluating an aligned language model by reading its answers assumes the answers carry the distinction the evaluator cares about. We introduce LLM end...
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.