495 citations · 521 across the 8 of their papers we have counts for
9 papers
Reasoning-Preserving Fine-Tuning of Post-RL LLMs with Null-Basis LoRA
Wenzhi Fang, Nicholas Tzou, Lazar Valkov +1
Reinforcement learning (RL)-based post-training has become an effective approach for eliciting reasoning capabilities in large language models (LLMs). However, adapting post-RL mod…
Not All Ranks Are Equal: Budget-Aware LoRA Merging Across Tasks
Avinash Amballa, Yashas Malur Saidutta, Wenbo Li +2
Merging low-rank adapters (LoRAs) promises to eliminate the overhead of swapping task-specific weights at inference time. However, existing merging methods assume every layer needs…
Doc-to-Atom: Learning to Compile and Compose Memory Atoms
Xingjian Diao, Wenbo Li, Yashas Malur Saidutta +3
Long input sequences are central to document understanding and multi-step reasoning in Large Language Models, yet the quadratic cost of attention makes inference both memory-intens…
Foveated Reasoning: Stateful, Action-based Visual Focusing for Vision-Language Models
Juhong Min, Lazar Valkov, Vitali Petsiuk +2
Vision-language models benefit from high-resolution images, but the increase in visual-token count incurs high compute overhead. Humans resolve this tension via foveation: a coarse…
Towards High-Fidelity Gaussian Splatting with Queried-Convolution Neural Networks
Abhinav Kumar, Tristan Aumentado-Armstrong, Lazar Valkov +4
Gaussian Splatting has revolutionized the field of Novel View Synthesis (NVS) with faster training and real-time rendering. However, its reconstruction fidelity still trails behind…
Adaptive Memory Replay for Continual Learning
James Seale Smith, Lazar Valkov, Shaunak Halbe +4
Foundation Models (FMs) have become the hallmark of modern AI, however, these models are trained on massive data, leading to financially expensive training. Updating FMs as new dat…
A Probabilistic Framework for Modular Continual Learning
Lazar Valkov, Akash Srivastava, Swarat Chaudhuri +1
Modular approaches that use a different composition of modules for each problem are a promising direction in continual learning (CL). However, searching through the large, discrete…
HOUDINI: Lifelong Learning as Program Synthesis
Lazar Valkov, Dipak Chaudhari, Akash Srivastava +2
We present a neurosymbolic framework for the lifelong learning of algorithmic tasks that mix perception and procedural reasoning. Reusing high-level concepts across domains and lea…
VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning
Akash Srivastava, Lazar Valkov, Chris Russell +2
Deep generative models provide powerful tools for distributions over complicated manifolds, such as those of natural images. But many of these methods, including generative adversa…