Papers by Andrey Kuznetsov

8 papers
Your Transformer is Secretly Linear (2024.acl-long)

Copied to clipboard

Challenge: a novel linear characteristic exclusive to transformer decoders is revealed: embedding transformations between sequential layers exhibit almost perfect linearity.
Approach: They propose a cosine-similarity-based regularization to reduce layer linearity in transformer decoders.
Outcome: The proposed method improves performance metrics on Tiny Stories and SuperGLUE but also decreases the linearity of the models.
Kandinsky 3: Text-to-Image Synthesis for Multifunctional Generative Framework (2024.emnlp-demo)

Copied to clipboard

Challenge: Text-to-image (T2I) diffusion models are popular for image manipulation, but also for video generation.
Approach: They propose a novel T2I diffusion model based on latent diffusion that extends the base model for various applications.
Outcome: The proposed model achieves high quality and photorealism and is 3 times faster than the base model.
The Shape of Learning: Anisotropy and Intrinsic Dimensions in Transformer-Based Models (2024.findings-eacl)

Copied to clipboard

Challenge: Embeddings in transformers encode vast amounts of linguistic nuances and patterns.
Approach: They investigate the anisotropy dynamics and intrinsic dimension of embeddings in transformers . they found that transformer decoders exhibit a bell-shaped anisotropie profile .
Outcome: The investigated embeddings exhibit a bell-shaped curve with the highest anisotropy concentrations in the middle layers . the intrinsic dimension increases in the initial phases of training, indicating an expansion into higher-dimensional space.
LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers (2025.findings-naacl)

Copied to clipboard

Challenge: Large Language Models (LLMs) encode and store contextual information, but internal mechanisms are opaque.
Approach: They propose a toolkit that assesses token-level nonlinearity, evaluates contextual memory, visualizes intermediate layer contributions and measures intrinsic dimensionality of representations.
Outcome: The proposed framework assesses token-level nonlinearity, evaluates contextual memory, visualizes intermediate layer contributions, and measures the intrinsic dimensionality of representations.
SPARTA: Evaluating Reasoning Segmentation Robustness through Black-Box Adversarial Paraphrasing in Text Autoencoder Latent Space (2026.eacl-long)

Copied to clipboard

Challenge: Existing work on semantically equivalent textual paraphrases has focused on perturbing image inputs.
Approach: They propose a novel adversarial paraphrasing task that generates grammatically correct paraphrases that sighed the original query meaning while degrading segmentation performance.
Outcome: The proposed task outperforms previous methods by up to 2x on ReasonSeg and LLMSeg-40k datasets.
Kandinsky: An Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion (2023.emnlp-demo)

Copied to clipboard

Challenge: Experimental evaluations demonstrate FID score of 8.03 on the COCO-30K dataset, marking our model as the top open source performer in terms of measurable image generation quality.
Approach: They propose a latent diffusion-based model that combines image prior and latent diffusive techniques to create a text-to-image architecture.
Outcome: The proposed model achieves the highest FID score among open-source models . it is compared with the state-of-the-art models on the COCO-30K dataset .
Fast and Accurate Fisher-Guided Quantization via Efficient Kronecker Factorization (2026.acl-long)

Copied to clipboard

Challenge: Quantization has shown strong results in preserving model quality under compression, but under aggressive bit-width reductions, even quantization may require additional information to prevent performance degradation.
Approach: They propose a Kronecker-factored approximation that captures second-order curvature information, captured by the Hessian, to achieve a 10 speedup over prior approaches.
Outcome: The proposed method significantly accelerates the most expensive component in second-order quantization – Hessian parameterization . it achieves up to a 10 speedup over prior approaches.
Bring the Apple, Not the Sofa: Impact of Irrelevant Context in Embodied AI Commands on VLA Models (2026.eacl-srw)

Copied to clipboard

Challenge: Embodied AI is undergoing rapid development, with robots increasingly exhibiting practical utility in everyday environments.
Approach: They evaluate the robustness of vision language action models under linguistic perturbations . they categorize irrelevant contexts into two groups according to their length and proximity to robot commands .
Outcome: The proposed model can exhibit relative robustness to random context, with a performance drop within 10%, the authors show . human paraphrases of instructions lead to a drop of nearly 20%, the study shows .

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations