VALC provides concept-level interpretations of FLMs, overcoming word-level limitations.
problem Lack of higher-level structure interpretation in FLMs' attention weights.
method Formal definition of conceptual interpretation, variational Bayesian framework (VALC).
result VALC finds optimal language concepts for FLM predictions, providing concept-level interpretations.
LORL learns object-centric representations from vision and language.
problem Learning disentangled, object-centric scene representations from vision and language.
method LORL integrates unsupervised object discovery and segmentation with language input to learn object-centric concepts.
result LORL improves unsupervised object discovery methods and aids downstream tasks.
Extends linear representation hypothesis to categorical and hierarchical concepts in LLMs.
problem Representing concepts without natural contrasts in large language models.
method Formalizes linear representation hypothesis for categorical and hierarchical concepts, proving relationships between concept hierarchy and representation geometry.
result Validated theoretical results on large language models, estimating representations for 900+ concepts.
The paper applies math and physics to language models, introducing entropy and geometric concepts.
problem Understanding and improving language models to approximate intelligent language.
method Formal definitions, functional analysis, topology, thermodynamics, and set theory.
result Entropy function reveals key obstacles for LLMs and offers insights into language models.
This work explains how linear representations in large language models arise from training objectives and gradient descent.
problem Understanding the origins of linear representations in large language models.
method A latent variable model to abstract and formalize concept dynamics, combined with analysis of the softmax cross-entropy objective and gradient descent.
result Linear representations emerge when learning from data matching the latent variable model, and this simple structure suffices to yield linear representations.
Unified model learns concepts across domains like left and right.
problem Limited generalization of language concepts in inference-only models.
method Logic-Enhanced Foundation Model (LEFT) with a differentiable, domain-independent program executor.
result LEFT flexibly learns and reasons with concepts across 2D images, 3D scenes, human motions, and robotic manipulation.
Paper introduces new graph concepts for better modeling of temporal interactions.
problem Graph theory struggles to capture temporal and structural aspects of interactions.
method Generalizes graph concepts to handle both temporal and structural aspects of interactions.
result Formalism allows direct modeling of interactions over time, similar to graph theory.
System solves a significant fraction of Bongard problems using visual features and pragmatic reasoning.
problem Solving Bongard problems with intelligent vision systems.
method Image processing, symbolic visual vocabulary, Bayesian inference, pragmatic reasoning.
result Good agreement between induced concepts and Bongard's solutions.
New approach extracts AI model representations for steering and monitoring.
problem Understanding and steering AI models' internal knowledge representation.
method Feature learning advances for scalable linear concept representations.
result Concept representations enable steering and monitoring of AI models.
Deep networks respond to specific linguistic units, not arbitrary patterns.
problem Understanding how deep convolutional networks interpret natural language.
method Concept alignment method based on unit responsiveness to replicated text.
result Deep networks selectively respond to morphemes, words, and phrases, not arbitrary patterns.
Paper proposes a streamlined approach to clinical concept extraction using LSTM-CRF.
problem Automated extraction of concepts from clinical records for clinical research.
method Bidirectional LSTM with CRF decoding initialized with general-purpose word embeddings.
result Experimental results outperform all recent methods and rank closely to the best submission from the original i2b2/VA challenge.
A new prior for language concepts inspired by consciousness neuroscience.
problem Learning representations of high-level language concepts.
method Inspired by cognitive neuroscience, a new prior is proposed to help disentangle abstract factors.
result A low-dimensional conscious state can make high-probability statements, consistent with sparse factor graphs.
This work improves interpretability in deep learning models by introducing a two-level concept discovery framework.
problem High complexity and lack of interpretability in deep learning models, especially for safety-critical tasks.
method Concept Bottleneck Models (CBMs) framework combining vision-language models and data-driven coarse-to-fine concept selection.
result The proposed framework outperforms recent CBM approaches and provides a principled interpretability.
Extends V-IP framework to use LLMs for generating task-relevant concepts, improving interpretability and performance.
problem Limited applicability of V-IP to small-scale tasks due to manual data annotation.
method Integrates Foundational Models with Large Language and Multimodal Models to generate and annotate concepts.
result FM+V-IP achieves better test performance with fewer concepts/queries compared to other frameworks.
Describes explaining neurons in deep representations using compositional logical concepts.
problem Interpreting neuron behavior in deep neural networks.
method Identifying compositional logical concepts that closely approximate neuron behavior.
result Compositional explanations provide insights into model performance and allow for adversarial example creation.
COCA refactors training data to identify and erase unsafe concepts in LLMs.
problem Identifying and erasing unsafe concepts in Large Language Models (LLMs) for safety alignment.
method Concept Concentration (COCA) refactors training data with an explicit reasoning process to identify and erase unsafe concepts.
result COCA significantly reduces both in-distribution and out-of-distribution jailbreak success rates while maintaining strong performance on regular tasks.
The task of associating images and videos with a natural language description has attracted a great amount of attention recently. Rapid progress has been made in terms of both developing novel algorithms and releasing new datasets. Indeed, the state-of-the-art results on some of the standard datasets have been pushed i…
Introduces Kundt spaces using geometric language.
problem Clarify properties of Kundt spaces.
method Geometric approach focusing on key concepts.
result Pedagogical introduction to Kundt spaces.
The study enhances financial rule matching using NLP without datasets.
problem Performing semantic matching between financial rules and policies.
method Outperforming pre-trained models with NLP techniques using free resources.
result Improved semantic matching between financial rules and policies.
Deep learning models generate languages that lack abstract reasoning.
problem Lack of abstract reasoning in deep learning-generated languages.
method Analyzed emergent language from two multi-agent games with compositional measures.
result Deep learning solutions often fail to generalize to out-of-training examples.
Introduces challenges and techniques for creating machine translation for indigenous languages.
problem Limited data for machine translation of indigenous languages.
method Introduction to challenges, concepts, and techniques for creating MT systems.
result Discussion of recent advances and open questions in NLP for these languages.
LLMs can be tricked into recalling facts based on context clues.
problem Manipulation of LLMs' factual recall through context changes.
method Mathematical exploration of transformers' associative memory properties.
result Transformers use self-attention and value matrix for associative memory.
The paper explores linear representations in language models using counterfactuals.
problem Understanding linear representations and geometric concepts in large language models.
method Formalized linear representation in output and input spaces, identified causal inner product.
result Unified understanding of linear representations and their connection to interpretation and control.
New bounds on query learning complexity for various concept classes.
problem Learning complexity in query learning models.
method Introducing new combinatorial quantities and proving lower and upper bounds.
result New and shorter proofs of efficient learnability for prominent examples.
Word2Vec embedding for Italian language developed.
problem Lack of word embedding for Italian language.
method Parameter tuning of Word2Vec for Italian language.
result An embedding for Italian language created.
Visual distance for WordNet synsets using deep learning features.
problem Measuring distances between concepts in WordNet.
method Extract visual features from ImageNet-trained CNNs and use them to represent synsets, defining a new distance measure.
result The proposed visual distance measure outperforms traditional lexical distances.
Quaternionic differential geometry expands geometric concepts using quaternions.
problem Generalizing geometric concepts to quaternionic constraints.
method Generalizing curves and surfaces, curvature, torsion, differential forms, and directional derivatives to quaternionic constraints.
result Quaternionic formalism provides a suitable language for differential geometry.
Measures faithfulness of LLM explanations to reveal hidden biases and misleading claims.
problem LLM explanations can misrepresent the model's reasoning process, leading to over-trust and misuse.
method Defines faithfulness in terms of concept influence and uses counterfactuals and Bayesian models to estimate it.
result Can quantify and discover interpretable patterns of unfaithfulness in LLM explanations.
Improves biomedical entity linking with latent type modeling.
problem Lack of fine-grained type information for entity disambiguation.
method Jointly models entity disambiguation and latent type learning without direct supervision.
result Significant performance improvements over state-of-the-art techniques.
Computer science scans LLMs to understand and manipulate their economic forecasts.
problem Understanding and controlling the reasoning of large language models in economics.
method Brain scanning techniques applied to LLMs to identify and manipulate underlying concepts.
result LLMs can be steered to generate forecasts with specific biases, allowing for correction or simulation.
Algorithm removes spurious concepts from neural network representations without harming task performance.
problem Spurious correlations hinder neural network out-of-distribution generalization.
method Iterative algorithm that identifies two orthogonal subspaces in neural network representation.
result Algorithm outperforms existing methods on computer vision and natural language processing benchmarks.
Proposes a method to align language and image data.
problem Aligning language and image data for better understanding.
method Uses triplet loss to learn consistent embeddings of language and images.
result Outperforms four baselines across multiple metrics.
The paper tests semantic importance in opaque models using betting.
problem Precise statistical guarantees for semantic concepts in black-box models.
method Formalizes global and local statistical importance via conditional independence and SKIT.
result Shows effectiveness and flexibility of the framework on various models.
Paper translates train track concepts to cluster algebras for pseudo-Anosov mapping classes.
problem Understanding pseudo-Anosov mapping classes on surfaces.
method Using Goncharov--Shen's potential function, the paper translates train track concepts into cluster algebra language.
result Proves sign stability of general pseudo-Anosov mapping classes.
Unified approach to learn interpretable concepts from data.
problem Building interpretable machine learning models and highly-performing foundation models.
method Relating causal representation learning and foundation models, defining concepts and proving their recoverability.
result Provable recovery of human-interpretable concepts from diverse data.
This research integrates attention into XAI frameworks for better model explanations.
problem Improving the interpretability of transformer models.
method Developed two novel explanation methods: Shapley value decomposition and token-level directional derivatives.
result Attention weights can be meaningfully incorporated into XAI frameworks, enhancing transformer explainability.
System helps engineers with concept recognition for SEVA.
problem Recognizing systems engineering concepts for SEVA.
method Token classification task, domain expert labeling, pre-trained model fine-tuning, essential datasets creation, knowledge graph construction.
result System successfully recognizes systems engineering concepts.
This paper introduces a method to find complete and interpretable concept-based explanations for deep neural networks.
problem Lack of complete and interpretable concept-based explanations in deep neural networks.
method Definition of completeness, concept discovery method, and importance score calculation using game-theoretic notions.
result The proposed method finds complete and interpretable concept-based explanations for deep neural networks.
A new framework for interpretable models using sparse linear layers.
problem Performance degradation and lower interpretability in concept bottleneck models.
method Contrastive Language Image models and a single sparse linear layer with Bayesian inference.
result Our framework outperforms recent CBM approaches in accuracy and concept sparsity.
New method extracts biological concepts from cell microscopy images.
problem Extracting meaningful concepts from vision foundation models trained on cell microscopy images.
method Sparse dictionary learning (DL) combined with PCA whitening pre-processing.
result Successfully retrieved biologically meaningful concepts like cell types and genetic perturbations.
Deep networks learn new words from few examples, like humans.
problem Deep learning requires large datasets to learn new concepts.
method Inspired by human learning, a simple technique to learn new words from little data.
result Deep networks can learn useful representations for new words from minimal data.
BC-LLM uses LLMs to find concepts without predefined sets, improving interpretability and performance.
problem Finding a balance between interpretability and accuracy in concept extraction models.
method Bayesian approach with LLMs as both concept extractor and prior.
result BC-LLM outperforms interpretable and black-box models across various datasets.
Researchers apply concept-based explainability to EEG data.
problem Understanding the internal states of complex EEG transformer models.
method Concept Activation Vectors (CAVs) adapted for EEG data, using externally labeled datasets and anatomically defined concepts.
result Both approaches to concept formation yield valuable insights into EEG model representations.
Geometric framework detects concept frustration between human concepts and machine representations.
problem Aligning human concepts with machine learning representations.
method Geometric framework and similarity measures for detecting concept frustration.
result Concept frustration affects machine learning model performance and reorganizes learned concept representations.
L3Ms fine-tune LLMs with constraints for tailored applications.
problem Inadequate alignment of LLMs for diverse applications.
method Formulate SFT and alignment as constrained optimization, using logarithmic barriers.
result Versatile and effective in achieving tailored alignments for various applications.
RNNs are crucial for text and speech tasks, explained in this overview.
problem Understanding RNNs for advanced AI tasks.
method Overview of RNN concepts and recent advances.
result Fundamentals and recent techniques in RNNs.
PCBMs turn any neural network into interpretable models without dense annotations.
problem Restrictive nature of CBMs and lack of dense concept annotations in training data.
method Introduce PCBMs that can turn any neural network into interpretable models without dense annotations.
result PCBMs can turn any neural network into interpretable models without dense annotations, improving interpretability and performance.
Survey of varifolds with focus on their regularity.
problem Understanding the concept and regularity of varifolds.
method Survey and examples of varifolds, emphasizing mean curvature and integral varifolds.
result Emphasis on the regularity of integral varifolds with mean curvature.