arXiv AI
Reference-Grafting Matches Fine-Tuning at Eliciting Sandbagged Capabilities
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Machine Learning
3d ago
A Causal Model for Locating and Unlocking Sandbagging in Model Organisms
arXiv:2608.29461v1 Announce Type: new Abstract: Sandbagging models strategically underperform on evaluations while retaining the capabilities being measured. The evaluations that guide frontier-model...
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.
arXiv Machine Learning
Jul 8
More Convincing, Not More Correct: Self-Play Reward Hacking of Reference-Free LLM Judges
arXiv:2607. 05904v1 Announce Type: new Abstract: Training a language model against its own reference-free judgments (the premise of self-rewarding, self-play, and LLM-as-a-judge pipelines) assumes a model's verdict on a shown answer tracks correctness.
arXiv AI
Jul 3
Procedural Memory Distillation: Online Reflection for Self-Improving Language Models
arXiv:2607. 01480v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR), along with recent selfdistillation variants such as SDPO, evaluates each rollout against a verifier and updates the policy from that episode-level signal.
arXiv Machine Learning
Aug 24
Nothing Changed but the Model: CellFill -- Bounded In-Cell Learning for Bit-Identical, Revocable Updates to Quantized LLMs
arXiv:2608.20873v1 Announce Type: new Abstract: Every way of teaching a deployed language model something new -- full fine-tuning, adapter merging, model editing -- replaces the released checkpoint,...
arXiv Machine Learning
Aug 27
Does Fine-Tuning Undo Activation Steering? Behavioural Recovery Without Weight-Edit Reversal
The paper investigates whether fine‑tuning a language model erases previously embedded activation steering interventions that suppress refusals and encourage brevity. Across five instruction‑tuned models (3B–14B) subjected to non‑adversarial supervised fine‑tuning (SFT) and reinforcement learning from human feedback (RLHF), the authors find that the steering’s behavioural effect degrades when the fine‑tuning objective conflicts with the targeted behaviour, yet the underlying weight edits remain largely unchanged. Mechanistically, the steering vectors survive with minimal alteration, but functionally the steering is vulnerable and must be re‑validated after downstream training.