Knowledge catalogue
arXiv:2607.00183v1 Announce Type: new Abstract: Adapting pre-trained text-to-image diffusion models, whether to learn new visual concepts or erase unwanted ones, is routinely evaluated on its intended
arXiv:2605.16362v1 Announce Type: cross Abstract: Activation steering offers a lightweight way to control LLMs without retraining, but its effectiveness varies sharply across concepts. Prior work ofte
arXiv:2605.08702v1 Announce Type: cross Abstract: This paper tackles compositional personalization of vision-language models (VLMs). In this problem, multiple user-defined concepts must be recognized
arXiv:2604.27759v1 Announce Type: cross Abstract: Integrating domain knowledge into deep neural networks is a promising way to improve generalization. Existing methods either encode prior knowledge in
arXiv:2502.06809v3 Announce Type: replace-cross Abstract: Pervasive polysemanticity in large language models (LLMs) undermines discrete neuron-concept attribution, posing a significant challenge for m
arXiv:2603.21396v2 Announce Type: replace Abstract: Recent work has shown that LLMs can sometimes detect when steering vectors are injected into their residual stream and identify the injected concept
arXiv:2604.08039v1 Announce Type: new Abstract: Interpreting the concepts encoded by individual neurons in deep neural networks is a crucial step towards understanding their complex decision-making pr
arXiv:2610.10379v1 Announce Type: new Abstract: Continual learning requires models to adapt to new domains and new classes while retaining prior knowledge. Many existing methods rely on continued opti
arXiv:2610.07638v1 Announce Type: new Abstract: As part of learning to play complex games, human players develop develop abstractions for concepts and strategies of gameplay consistent with game rules
arXiv:2609.37537v1 Announce Type: new Abstract: Machine unlearning has emerged as a critical post-hoc safety measure to erase sensitive concepts from Text-to-Image (T2I) models without prohibitive ret
Hi again! This is a update to my previous post MageTrail V0.25 where I continued this tech demo toy model thingy called MageTrail, a Danbooru/E621 proof of concept Full-Finetune of Microsoft's MageFlo
arXiv:2609.35269v1 Announce Type: cross Abstract: The rising number of concept unlearning techniques for text-to-image (T2I) diffusion models has produced a fragmented evaluation landscape. Methods ar
arXiv:2609.32355v1 Announce Type: cross Abstract: Model-internal features can be studied through both their ability to identify a specified concept and their causal effect when manipulated, e.g., thro
arXiv:2609.16283v1 Announce Type: new Abstract: exttt{URANIA} provides end-to-end differential privacy (DP) for summaries of data-dependent clusters. However, its cluster--keyword release does not dir
arXiv:2608.27031v1 Announce Type: new Abstract: Measuring alignment between documents and structured reference frameworks requires identifying conceptual evidence distributed throughout the text and r
arXiv:2608.23065v1 Announce Type: cross Abstract: Cultural understanding in video means more than recognizing what is visible; it requires grasping the symbolic and temporal significance of cultural c
arXiv:2608.02957v1 Announce Type: new Abstract: Steering vectors (SVs) are widely used to influence the expression of concepts (e.g., truthfulness) in large language model outputs. A key assumption un
Back in 2024 I tweeted screenshots of a game concept generated by GPT-3 and some concept 'art' created using DALL-E. Today, on the fourth anniversary of that tweet, I decided to see if Claude Fable 5
TLDR: There is no one prompting trick that will result in Krea 2 Turbo giving you exactly the style you want and across the whole image. Instead, if you are trying to achieve styles without the use of
arXiv:2607.21491v1 Announce Type: new Abstract: Do independently trained language models come to represent the same thing in the same way? We answer for code, extending a recently introduced concept-c
arXiv:2606.05409v1 Announce Type: cross Abstract: Vision-language models (VLMs), like human learners, are frequently exposed to new visual concepts, but how they map novel visual references to languag
arXiv:2512.08560v3 Announce Type: replace Abstract: Understanding how the human brain represents visual concepts, and in which brain regions these representations are encoded, remains a long-standing
arXiv:2602.03410v2 Announce Type: replace Abstract: Recent advances in large-scale diffusion models have intensified concerns about their potential misuse, particularly in generating realistic yet har
arXiv:2605.04980v1 Announce Type: cross Abstract: Activation-based steering provides control of LLM behavior at inference time, but the dominant paradigm reduces each concept to a single direction who
arXiv:2604.22676v1 Announce Type: new Abstract: Graph neural networks achieve strong node-classification accuracy, but their learned message passing entangles ego attributes, neighborhood smoothing, h
arXiv:2512.05794v2 Announce Type: replace-cross Abstract: Sparse autoencoders (SAEs) are a mechanistic interpretability technique that have been used to provide insight into learned concepts within la
arXiv:2512.04585v3 Announce Type: replace Abstract: Segment Anything Model 3 (SAM3) advances open-vocabulary segmentation through promptable concept segmentation, enabling users to segment all instanc
arXiv:2610.10594v1 Announce Type: new Abstract: Activation probes that monitor deployed language models are trained on synthetic conversations, and how many a probe needs is open. We trace learning cu
arXiv:2610.05853v1 Announce Type: new Abstract: Monitoring concept drift from an adaptive classifier's error stream creates an operational conflict with the model's own update loop. When internal adap
arXiv:2603.10210v2 Announce Type: replace-cross Abstract: While Diffusion Models excel in text-to-image synthesis, they frequently suffer from catastrophic concept omission when generating complex mul
arXiv:2609.38325v1 Announce Type: new Abstract: We introduce modal kinetic typography, which animates a vector glyph to express a semantic concept while keeping it legible. Our key idea is to build mo
arXiv:2511.17735v2 Announce Type: replace Abstract: Foundation models in several scientific domains, including visual domains, learn representations that capture complex semantics from their respectiv
arXiv:2609.26704v1 Announce Type: new Abstract: Large language models increasingly tackle hard reasoning problems by spending more test-time compute, yet the dominant strategy remains naive repeated s
arXiv:2609.10723v1 Announce Type: new Abstract: Text-to-image diffusion transformers (DiTs) are powerful generators, yet direct prompting provides limited control interface for style intensity and can
arXiv:2608.13630v1 Announce Type: new Abstract: We present ConceptFlow, a scikit-learn-compatible Python library for Formal Concept Analysis that constructs and renders nested line diagrams from many-
arXiv:2512.02657v2 Announce Type: replace-cross Abstract: Real-world deployment of text-to-image diffusion models requires continual concept removal as new privacy, copyright, or safety obligations ar
arXiv:2606.29909v1 Announce Type: new Abstract: Encrypted traffic classification has achieved strong performance, but its decision process remains difficult to interpret. Existing methods usually comb
arXiv:2606.01215v1 Announce Type: cross Abstract: Current 3D spatial reasoning methods face a fundamental trade-off: neuro-symbolic 3D (NS3D) concept learners achieve interpretable reasoning through c
arXiv:2606.02129v1 Announce Type: new Abstract: Image customization learns target subjects from reference concept images and generates conditioned images per text prompts, mainly modifying styles or b
arXiv:1607.06330v2 Announce Type: replace Abstract: In this doctoral thesis, we apply premises of cognitive linguistics to terminological definitions and present a proposal called the flexible termino
arXiv:2605.00474v1 Announce Type: new Abstract: Modern vision models achieve remarkable accuracy, but explaining where evidence arises, what the model encodes, and how internal computations assemble t
arXiv:2506.16742v3 Announce Type: replace Abstract: To be adopted in safety-critical domains like medical image analysis, AI systems must provide human-interpretable decisions. Variational Information
arXiv:2604.23377v1 Announce Type: new Abstract: Neurosymbolic systems can satisfy logical constraints during learning without achieving the intended concept-label correspondence; this is a problem kno
arXiv:2511.16698v2 Announce Type: replace Abstract: SNOMED CT is a biomedical ontology with a hierarchical representation, modelling terminological concepts at a large scale. Knowledge retrieval in SN
arXiv:2610.03621v1 Announce Type: cross Abstract: There has been a surge of recent work on correlated equilibrium concepts in Markov games. However, existing results focus on concepts weaker than norm
Jev from TypeSafe AI got me thinking about non-autoregressive decision engines, so I built a small proof-of-concept called AES to test the concept locally on open weights. It uses Qwen/Qwen2.5-1.5B-In
arXiv:2609.10055v1 Announce Type: cross Abstract: Biomedical ontology normalization maps free-text expressions to standardized concepts, enabling consistent integration and analysis of biomedical data
Hi! Today I'm releasing MageTrail, a Danbooru/E621 proof of concept Full-Finetune of Microsoft's MageFlow 4B T2I model, using a diversity maximized condensed 41k images dataset (originally made by Lod
arXiv:2609.00003v1 Announce Type: new Abstract: Machine unlearning studies the removal of knowledge from an AI model, making the system forget a concept it previously learned. Despite rapid progress i
arXiv:2608.30130v1 Announce Type: cross Abstract: Retrieval-augmented language models can fail to respect negative constraints when the retriever supplies evidence about concepts the user explicitly e
arXiv:2607.16325v2 Announce Type: replace Abstract: Foundation models provide powerful representations for brain MRI analysis, but their predictions remain difficult to interpret in anatomically meani
arXiv:2608.25876v1 Announce Type: cross Abstract: Generative design interfaces increasingly expose semantic controls that let users steer output with concepts such as 'more elegant' or 'more minimalis
arXiv:2607.28936v1 Announce Type: cross Abstract: Facial biometric identification relies on the distinctiveness of user attributes within a high-dimensional embedding space. However, the decision boun
arXiv:2607.25132v1 Announce Type: new Abstract: A central challenge in interpreting learned decision-making systems is to determine whether their internal representations contain concepts that help ex
arXiv:2607.12886v1 Announce Type: new Abstract: Clinical notes contain many of the signs and symptoms that bring patients to care, yet this information rarely reaches structured fields. Existing extra
arXiv:2607.04800v1 Announce Type: new Abstract: Neural networks are thought to represent concepts as directions in their activation space, and superposition lets them encode more concepts than they ha
arXiv:2606.31699v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) have recently been proposed as interpretable tools for concept-level manipulation, under the assumption that isolated featu
arXiv:2606.08311v1 Announce Type: new Abstract: Electronic health record (EHR) notes are dense medical documents containing large amounts of information, often filled with complex medical jargon. High
arXiv:2606.03245v1 Announce Type: cross Abstract: Concepts of calibration formalize the compatibility between probabilistic predictions and the respective outcomes. In a nutshell, the outcomes ought t
arXiv:2605.30825v1 Announce Type: cross Abstract: Unlearning in diffusion models aims to remove undesirable data or concepts while preserving the utility of pretrained models -- two fundamentally conf