Paper Bites
Each Paper Bite is a focused Markdown note: why the paper matters, the core idea, how it works, what to inspect in the results, and practical takeaways.
MMR-AD builds a large multimodal industrial anomaly benchmark with reasoning traces, localization annotations, and cross-dataset evaluation splits, then uses it to show that current generalist MLLMs still struggle badly on practical anomaly detection and especially anomaly localization.
MoECLIP adapts CLIP to zero-shot anomaly detection by routing each patch through specialized LoRA experts, then explicitly separating those experts so patch-level specialization improves without giving up CLIP's transfer ability.
ThinkSound treats video-to-audio as a reasoning problem rather than a one-shot mapping, using multimodal chain-of-thought to guide a unified model through foley generation, object-focused refinement, and instruction-based editing.
DreamBooth turns subject-driven image generation into a lightweight personalization problem: bind a rare token to a specific subject, then regularize fine-tuning so the model keeps both the subject identity and the broader class prior.
SEAL improves iterative training by perturbing early layers with short bursts of gradient ascent instead of resetting later layers, leading to better in-distribution accuracy and much stronger transfer behavior than LLF.
Lance argues that a unified image-video model works better when tasks share context but not every representation or parameter path, combining multi-task training with separate understanding and generation routes.
Ditto turns instruction-based video editing into a data problem first: it builds a million-sample synthetic dataset, then trains Editto with curriculum learning to follow text edits more reliably.
Chain-of-Zoom treats extreme super-resolution as a sequence of zoom steps, using multi-scale prompts and preference alignment to guide details beyond a model's usual scale range.
LoRA freezes a pretrained language model and learns small low-rank update matrices, making task adaptation much cheaper to train, store, and switch.
ControlNet adds spatial conditioning to pretrained text-to-image diffusion models while protecting the original model's generation quality through zero-initialized control branches.
ViT-CoMer keeps the flexibility of a plain Vision Transformer, then adds a parallel convolutional branch so dense prediction tasks can use richer multi-scale features.