13 papers
INTACT: Inducing Latent Reasoning on a Frozen Backbone
Kutay Tire, Atula Tejaswi, Haris Vikalo +1
Latent reasoning replaces natural-language chains of thought with compact continuous "soft-token" representations, reducing the number of intermediate tokens that must be decoded a…
JOVE: Joint Execution and Verification for Resource-Aware LLM Task Graphs
Haoran Zhang, Dongjun Kim, Seohyeon Cha +4
Complex reasoning queries can be decomposed into directed acyclic task graphs and distributed across heterogeneous LLMs, reducing latency through parallelism and enabling smaller m…
Twist, Don't Tilt: Trajectory-Exact Constrained Decoding for Masked Diffusion Models
Aditya Thimmaiah, Lara Marinov, Jayanth Srinivasa +3
Constrained decoding for Masked Diffusion Language Models (MDLMs) aims to ensure that generated outputs satisfy a specified structure or syntax constraint. MDLMs generate outputs b…
Foundation-Preserving Optimization in Generalized Eigenspace
Dongjun Kim, Adrian de Wynter, Huancheng Chen +2
While finetuning effectively adapts foundation models to specialized downstream tasks, it can degrade nontarget capabilities acquired during pretraining. Existing forgetting aware…
Your Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systems
Jianing Zhu, Yeonju Ro, John Robertson +5
Long-lived AI agents are increasingly deployed as persistent operational systems, yet they are still evaluated like freshly initialized models. Day-one benchmarks miss a basic syst…
When Is Rank-1 Steering Cheap? Geometry, Granularity, and Budgeted Search
John T. Robertson, Jianing Zhu, Haris Vikalo +1
Activation steering offers a lightweight way to control LLMs without retraining, but its effectiveness varies sharply across concepts. Prior work often reads this variability as ev…
Online Learning for Multi-Layer Hierarchical Inference under Partial and Policy-Dependent Feedback
Haoran Zhang, Seohyeon Cha, Hasan Burhan Beytur +3
Hierarchical inference systems route tasks across multiple computational layers, where each node may either finalize a prediction locally or offload the task to a node in the next…
FedRot-LoRA: Mitigating Rotational Misalignment in Federated LoRA
Haoran Zhang, Dongjun Kim, Seohyeon Cha +1
Federated LoRA provides a communication-efficient mechanism for fine-tuning large language models on decentralized data. In practice, however, a discrepancy between the factor-wise…
CoreQ: Learning-Free Mismatch Correction and Successive Rounding for Quantization
Seohyeon Cha, Huancheng Chen, Dongjun Kim +4
Post-training quantization (PTQ) enables efficient deployment of large language models by mapping pretrained weights to low-bit formats without retraining, typically using a small…
Robust Super-Capacity SRS Channel Inpainting via Diffusion Models
Usman Akram, Fan Zhang, Yang Li +1
Accurate channel state information (CSI) is essential for reliable multiuser MIMO operation. In 5G NR, reciprocity-based beamforming via uplink Sounding Reference Signals (SRS) fac…
Federated Self-Supervised Modulation Classification under Non-IID and Imbalanced Data
Usman Akram, Yiyue Chen, Haris Vikalo
Automatic modulation classification (AMC) is a core enabler of cognitive wireless systems, providing spectrum awareness and supporting adaptive communication at the network edge. H…
Fed-REACT: Federated Representation Learning for Heterogeneous and Evolving Data
Yiyue Chen, Usman Akram, Chianing Wang +1
Motivated by the high resource costs and privacy concerns associated with centralized machine learning, federated learning (FL) has emerged as an efficient alternative that enables…
Task-Agnostic Federated Continual Learning via Replay-Free Gradient Projection
Seohyeon Cha, Huancheng Chen, Haris Vikalo
Federated continual learning (FCL) enables collaborative model training across distributed clients on sequentially arriving tasks without revisiting past data. However, existing ap…