Large Language Models (LLMs) deployed in real-world settings increasingly face the need to unlearn sensitive, outdated, or proprietary information. Existing unlearning methods typically formulate forgetting and retention as a regularized trade-off, combining both objectives into a single scalarized loss. This often leads to unstable optimization and degraded performance on retained data, especially under aggressive forgetting. We propose a new formulation of LLM unlearning as a constrained optimization problem: forgetting is enforced via a novel logit-margin flattening loss that explicitly drives the output distribution toward uniformity on a designated forget set, while retention is preserved through a hard constraint on a separate retain set. Compared to entropy-based objectives, our loss is softmax-free, numerically stable, and maintains non-vanishing gradients, enabling more efficient and robust optimization. We solve the constrained problem using a scalable primal-dual algorithm that exposes the trade-off between forgetting and retention through the dynamics of the dual variable, all without any extra computational overhead. Evaluations on the TOFU and MUSE benchmarks across diverse LLM architectures demonstrate that our approach consistently matches or exceeds state-of-the-art baselines, effectively removing targeted information while preserving downstream utility.

Organisms constantly pivot between tasks such as evading predators, foraging, traversing rugged terrain, and socializing, often within milliseconds. Remarkably, they preserve knowledge of once-learned environments sans catastrophic forgetting, a phenomenon neuroscientists hypothesize, is due to a singular neural circuitry dynamically overlayed by neuromodulatory agents such as dopamine and acetylcholine. In parallel, deep learning research addresses analogous challenges via domain generalization ($\textbf{DG}$) and continual learning ($\textbf{CL}$), yet these methods remain siloed, despite the brain's ability to perform them seamlessly. In particular, prior work has not explored architectures involving associative memories ($\textbf{AM}$s), which are an integral part of biological systems, to jointly address these tasks. We propose Memory-Integrated Reconfigurable Adapters ($\textbf{MIRA}$), a unified framework that integrates Hopfield-style associative memory modules atop a shared backbone. These memory modules store adapter-weight updates as values and retrieve them via learned keys. Associative memory keys are learned post-hoc to index and retrieve an affine combination of stored adapter updates for any given task or domain on a per-sample basis. By varying only the task-specific objectives, we demonstrate that $\textbf{MIRA}$ seamlessly accommodates domain shifts and sequential task exposures under one roof. Empirical evaluations on standard benchmarks confirm that our $\textbf{AM}$-augmented architecture significantly enhances adaptability and retention: in $\textbf{DG}$, $\textbf{MIRA}$ achieves SoTA out-of-distribution accuracy, and in incremental learning settings, it outperforms architectures explicitly designed to handle catastrophic forgetting using generic $\textbf{CL}$ algorithms.

Continual learning of partially similar tasks poses a challenge for artificial neural networks, as task similarity presents both an opportunity for knowledge transfer and a risk of interference and catastrophic forgetting.However, it remains unclear how task similarity in input features and readout patterns influences knowledge transfer and forgetting, as well as how they interact with common algorithms for continual learning.Here, we develop a linear teacher-student model with latent structure and show analytically that high input feature similarity coupled with low readout similarity is catastrophic for both knowledge transfer and retention. Conversely, the opposite scenario is relatively benign. Our analysis further reveals that task-dependent activity gating improves knowledge retention at the expense of transfer, while task-dependent plasticity gating does not affect either retention or transfer performance at the over-parameterized limit. In contrast, weight regularization based on the Fisher information metric significantly improves retention, regardless of task similarity, without compromising transfer performance. Nevertheless, its diagonal approximation and regularization in the Euclidean space are much less robust against task similarity. We demonstrate consistent results in a permuted MNIST task with latent variables. Overall, this work provides insights into when continual learning is difficult and how to mitigate it.

However, practical model deployment often operates in the gap between these two limit cases, as real-world applications demand adaptation to specific subdomains, tasks or concepts -- spread over the entire, varying life cycle of a model.

Our technique greatly reduces the appearance of banding, without requiring any additional computation or post-processing at reconstructiontime.

However, as the number of serving tokens increases, the key-value (KV) caches occupy a lot of GPU memory, rendering the inference of large language models memory-bounded [29].

Vision-language models (VLMs) have transformed multimodal reasoning, but feeding hundreds of visual patch tokens into LLMs incurs quadratic computational costs, straining memory and context windows. Traditional approaches face a trade-off: continuous compression dilutes high-level semantics such as object identities, while discrete quantization loses fine-grained details such as textures. We introduce HTC-VLM, a hybrid framework that disentangles semantics and appearance through dual channels, i.e., a continuous pathway for fine-grained details via ViT patches and a discrete pathway for symbolic anchors using MGVQ quantization projected to four tokens. These are fused into a 580-token hybrid sequence and compressed into a single voco token via a disentanglement attention mask and bottleneck, ensuring efficient and grounded representations. HTC-VLM achieves an average performance retention of 87.2 percent across seven benchmarks (GQA, VQAv2, MMBench, MME, POPE, SEED-Bench, ScienceQA-Image), outperforming the leading continuous baseline at 81.0 percent with a 580-to-1 compression ratio. Attention analyses show that the compressed token prioritizes the discrete anchor, validating its semantic guidance. Our work demonstrates that a minimalist hybrid design can resolve the efficiency-fidelity dilemma and advance scalable VLMs.

Vision-based tactile sensors (VBTS) face a fundamental trade-off in marker and markerless design on the tactile skin: opaque ink markers enable measurement of force and tangential displacement but completely occlude geometric features necessary for object and texture classification, while markerless skin preserves surface details but struggles in measuring tangential displacements effectively. Current practice to solve the above problem via UV lighting or virtual transfer using learning-based models introduces hardware complexity or computing burdens. This paper introduces MagicSkin, a novel tactile skin with translucent, tinted markers balancing the modes of marker and markerless for VBTS. It enables simultaneous tangential displacement tracking, force prediction, and surface detail preservation. This skin is easy to plug into GelSight-family sensors without requiring additional hardware or software tools. We comprehensively evaluate MagicSkin in downstream tasks. The translucent markers impressively enhance rather than degrade sensing performance compared with traditional markerless and inked marker design: it achieves best performance in object classification (99.17\%), texture classification (93.51\%), tangential displacement tracking (97\% point retention) and force prediction (66\% improvement in total force error). These experimental results demonstrate that translucent skin eliminates the traditional performance trade-off in marker or markerless modes, paving the way for multimodal tactile sensing essential in tactile robotics. See videos at this \href{https://zhuochenn.github.io/MagicSkin_project/}{link}.