arXiv AI
Aug 7
Subliminal Learning is Non-Semantic Distillation
arXiv:2608. 05734v1 Announce Type: new Abstract: Subliminal Learning (SL) is a surprising type of generalization displayed by modern language models.
Subliminal Learning Is Steering Vector Distillation
arXiv:2606. 00995v1 Announce Type: new Abstract: Subliminal learning refers to a student language model acquiring a teacher's traits (e.
Related stories
Hugging Face Trending Papers
Aug 6
Subliminal Learning is Non-Semantic Distillation
Subliminal Learning (SL) is a surprising type of generalization displayed by modern language models. It allows the transfer of a bias or behavior from a teacher model to a student by distilling from seemingly unrelated or random synthetic data from the teacher.
arXiv Machine Learning
Sep 22
On Mitigation of Subliminal Learning in Large Language Models
arXiv:2609.22215v1 Announce Type: cross Abstract: Knowledge distillation can transmit unintended behavioral traits from a teacher model to a student through training data that appear semantically unr...
arXiv AI
Sep 16
Verbalizing Subliminal Learning Effects Using Text Optimization
The paper introduces SALVE, a method that uses text optimization to uncover and verbalize subliminal learning effects in language models. By optimizing a soft prompt and converting it into a legible text prompt, SALVE can reliably recover the teacher model’s hidden traits that are transmitted through a distillation dataset. The authors demonstrate SALVE’s effectiveness across various scenarios, including mixed datasets, biased teacher activation, and preference‑selected data, thereby providing a tool for detecting hidden influences in model training.
arXiv Machine Learning
Sep 2
Subliminal Learning as Trait-Direction Drift: A Mechanism and Targeted Control under SFT Distillation
The paper investigates subliminal learning, where hidden traits from a teacher model are transferred to a student during distillation. It introduces trait‑direction drift as the underlying mechanism, showing that biased generation creates measurable preference gaps that accumulate into behavioral transfer during fine‑tuning. The authors propose probe‑space corridor regularization, a targeted defense that constrains drift along a calibrated trait direction, significantly reducing hidden‑trait transfer while maintaining task performance.
arXiv AI
4d ago
Beyond Owls: Subliminal Learning Can Transfer Learned Capabilities and Backdoors
The paper investigates whether subliminal learning (SL) can transfer a broader set of traits beyond simple preferences, such as complex capabilities, backdoors, and hacking tendencies. Experiments show that SL can indeed impart a student model with the ability to predict outputs of a random MLP, partially embed a backdoor that triggers French responses to female names, and increase hacking success in a chess environment. The effectiveness of transfer varies with the distillation method, being enhanced by logit distillation or limiting LoRA to attention layers.
arXiv AI
1d ago
Defence against the dark arts of distillation: Unsupervised discovery of subliminal learning
The paper presents a method for detecting and mitigating subliminally transferred traits in language model distillation when the teacher and student share base parameters. By analyzing parameter updates across diverse datasets, the authors recover a weight‑space direction that reveals hidden traits such as animal preferences, without requiring trait labels. This direction can then be amplified for interpretation or projected out to suppress the trait with minimal impact on task performance.
arXiv AI
Sep 18
Can Data Attribution Filter Out Subliminal Learning? Not Reliably
The paper investigates whether gradient‑based data attribution methods can filter out subliminal learning in language models. Three attribution techniques—GradCos, a contrastive variant, and EK‑FAC—are evaluated on three models and compared to divergence tokens, a strong baseline that requires counterfactual teacher models. Results show that EK‑FAC can mitigate some subliminal effects at the token level, but overall the methods are inconsistent and generally less effective than divergence tokens, especially when filtering entire samples.
arXiv AI
Jul 13
GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs
arXiv:2507. 18043v2 Announce Type: replace-cross Abstract: Inference-time steering methods offer a lightweight alternative to fine-tuning large language models (LLMs) and vision-language models (VLMs) by modifying internal activations at test time without updating model weights.
arXiv AI
Aug 3
On the Generalization of Steering Vectors for Chain-of-Thought Faithfulness
arXiv:2607. 29062v1 Announce Type: new Abstract: Model capabilities have improved in large part due to scaling chain of thought.
arXiv AI
Sep 30
Signatures of semantic search in the activations of large language models
arXiv:2609.35599v2 Announce Type: replace Abstract: When recalling lists of concepts (e.g., animals) during the semantic fluency task (SFT), both humans and large language models (LLMs) organise thei...
arXiv AI
Aug 11
DeltaPrompts: Escaping the Zero-Delta Trap in Multimodal Distillation
arXiv:2605. 15532v3 Announce Type: replace-cross Abstract: Distillation enables compact Vision-Language Models (VLMs) to obtain strong reasoning capabilities, yet the prompts driving this process are typically chosen via simple heuristics or aggregated from off-the-shelf datasets.