The University of Cambridge has won its fight to stop a rowing company based in the city trademarking its name. It argued Cambridge Rowing Limited would be able to take unfair advantage of and cause detriment to the university's reputation if its logo was registered. The university owns trademarks for the word Cambridge, meaning it has the right to stop others from using it in certain circumstances. Omar Terywall, the company's founder, said he was gutted at the outcome and the case had been a terrifying ordeal. He said he hoped to appeal the decision by the Intellectual Property Office (IPO).

We thank all the anonymous reviewers for their constructive feedback. We address each comment as follows. R1-Q2:Just using the predicted mask to concat. R1-Q3:Refine the predicted mask with CRF . SEAM show that CRF ( vs CONT A) is only effective in the first round, i .

We aim to understand grokking, a phenomenon where models generalize long after overfitting their training set. We present both a microscopic analysis anchored by an effective theory and a macroscopic analysis of phase diagrams describing learning performance across hyperparameters. We find that generalization originates from structured representations, whose training dynamics and dependence on training set size can be predicted by our effective theory (in a toy setting). We observe empirically the presence of four learning phases: comprehension, grokking, memorization, and confusion. We find representation learning to occur only in a Goldilocks zone (including comprehension and grokking) between memorization and confusion. Compared to the comprehension phase, the grokking phase stays closer to the memorization phase, leading to delayed generalization. The Goldilocks phase is reminiscent of intelligence from starvation in Darwinian evolution, where resource limitations drive discovery of more efficient solutions. This study not only provides intuitive explanations of the origin of grokking, but also highlights the usefulness of physics-inspired tools, e.g., effective theories and phase diagrams, for understanding deep learning.

Diabetic retinopathy (DR) grading plays a critical role in early clinical intervention and vision preservation. Recent explorations predominantly focus on visual lesion feature extraction through data processing and domain decoupling strategies. However, they generally overlook domain-invariant pathological patterns and underutilize the rich contextual knowledge of foundation models, relying solely on visual information, which is insufficient for distinguishing subtle pathological variations. Therefore, we propose integrating fine-grained pathological descriptions to complement prototypes with additional context, thereby resolving ambiguities in borderline cases. Specifically, we propose a Hierarchical Anchor Prototype Modulation (HAPM) framework to facilitate DR grading. First, we introduce a variance spectrum-driven anchor prototype library that preserves domain-invariant pathological patterns. We further employ a hierarchical differential prompt gating mechanism, dynamically selecting discriminative semantic prompts from both LVLM and LLM sources to address semantic confusion between adjacent DR grades. Finally, we utilize a two-stage prototype modulation strategy that progressively integrates clinical knowledge into visual prototypes through a Pathological Semantic Injector (PSI) and a Discriminative Prototype Enhancer (DPE). Extensive experiments across eight public datasets demonstrate that our approach achieves pathology-guided prototype evolution while outperforming state-of-the-art methods. The code is available at https://github.com/zhcz328/HAPM.

We propose Confusions over Time (CoT), a novel generative framework which facilitates a multi-granular analysis of the decision making process. The CoT not only models the confusions or error properties of individual decision makers and their evolution over time, but also allows us to obtain diagnostic insights into the collective decision making process in an interpretable manner.

Widely adopted in modern Vision Transformer designs, Softmax attention can effectively capture long-range visual information; however, it incurs excessive computational cost when dealing with high-resolution inputs.

I do not completely understand (apart for some parts of the proofs) why refer to these functions as Graph-based. Boolean k-ary functions may be thought of as hyper-graphs. The definition shouldn't be unusual and it will be clarified to avoid any possible This is completely analogous to the standard empirical distribution for hypotheses classes. It might be helpful to summarise, ..., some basic properties of this new notion of VC dimension... ..., is there a Sauer-Shelah type upper bound on the size of the class in terms of the graph VC dimension? VC dimension entail small graph VC dimension). Shelah Lemma for graph VC dimension, indeed this is noteworthy and we should discuss this in the main text.

We thank all the reviewers' insightful suggestions. We will add SoT A numbers in the final version as R3 suggested. We will remove it in the final version. We've updated Figure 4 (a) using the new Figure 6 (in the Appendix), we see that the average number of iterations grows slowly with the sentence length. Why learning from teacher is better than oracle?

We sincerely thank all the reviewers for their insightful suggestions. We will add them back in the updated version, which will have 1 more page. Finally, we relax BERT and fine-tune the two models jointly. Results are shown in Table 1. As can be seen, this auxiliary training objective introduces a +0.8 F1 performance boost.

We thank the reviewers for their thoughtful comments and feedback. Below we respond to the reviewers' concerns. In the proof of Lemma A.5, on lines 429-430, we are using the Taylor expansion of We plan to expand our experimental results in multiple directions. 1) We have already If accepted, we plan to include it in the final submission. We apologize for the confusion. Note that the use of SGD or Adam does not change the overall takeaways of the experiments.