arXiv AI
Local Evidence and Geometric Readout Repair in Trained GNNs
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Sep 1
Forget or Fine-tune? A Comparative Study of Machine Unlearning Strategies for Noisy Label Correction
The paper compares five machine unlearning (MU) methods—NegGrad, Fine‑Tuning (FT), Random Labeling (RL), SalUn, and MUNBa—on noisy‑label correction across CIFAR‑10, CIFAR‑100, and Food‑101N. Results show that the best MU strategy depends on the noise type: FT works well for most closed‑set noise, RL and SalUn are robust and nearly match retraining accuracy under instance‑dependent noise, while MUNBa excels only under extreme symmetric noise. In open‑set noise, retraining on the cleaned data actually hurts performance, indicating that approximating retraining is not suitable in that regime, yet all MU methods still achieve near‑retraining accuracy on Food‑101N with much lower runtime.
arXiv Machine Learning
Sep 7
PACE: Propagation-Aware Collaborative Correction for One-Shot Personalized Federated Graph Learning
PACE introduces a propagation‑aware collaborative correction for one‑shot personalized federated graph learning. Each client sends a rank‑r update and a diagonal sketch of message moments, allowing the server to construct a correction that anchors to the receiver’s local model. A convex negative‑log‑likelihood calibration selects a single coefficient to blend local and external logits, improving accuracy and weighted‑F1 on most datasets while preserving local predictions when the correction is unhelpful.
arXiv AI
Sep 15
The Misery of Mechanistic Interpretability: A Formal Perspective
arXiv:2609.15533v1 Announce Type: cross Abstract: Mechanistic interpretability has become the dominant lens for understanding frontier language models, as their inner workings are complex and inheren...
arXiv AI
Jul 29
Why Does Grounding Hurt Medical VQA? Benchmarking, Diagnosis, and Fine-Tuning of Vision-Language Models
arXiv:2604. 27720v2 Announce Type: replace Abstract: Vision-language models (VLMs) are increasingly applied to medical visual question answering (Med-VQA), yet whether they can \emph{localize} the evidence behind their answers---a prerequisite for clinical auditability---is poorly characterized.
arXiv Machine Learning
Sep 14
Dead Weights, Live Signals: Feedforward Graphs of Frozen Language Models
The paper introduces a feedforward graph architecture that uses several frozen large language models as computational nodes connected through a shared continuous latent space via learned linear projections. By jointly optimizing projection matrices through backpropagation, the system combines the representations of three small frozen models with two larger ones, culminating in a lightweight cross‑attention output node. With only 17.6 M trainable parameters, the architecture attains state‑of‑the‑art results on ARC‑Challenge, OpenBookQA, and MMLU, surpassing both individual constituent models and parameter‑matched learned classifiers.
arXiv Machine Learning
Aug 27
Distance Is Not Enough: Forget-Retain Alignment Gap Predicts LLM Relearning Robustness
The paper introduces the Forget‑Retain Alignment Gap (FRAG), a training‑free metric that evaluates how well an update to a large language model (LLM) aligns with the principle of affecting forget‑critical weights while sparing retain‑critical ones. Unlike traditional robustness predictors that rely on global weight‑space displacement, FRAG distinguishes selective from dense updates and predicts relearning robustness without running a relearning attack. The authors also propose Forget‑Retain Pruning (FRP), which leverages this principle to enhance the robustness of unlearning in LLMs.