SteeringDiffusion supplies a bottlenecked, prompt-conditioned activation interface for frozen diffusion models that delivers smooth monotonic content-style control via one runtime scalar and timestep gating.
Title resolution pending
16 Pith papers cite this work. Polarity classification is still indexing.
16
Pith papers citing it
citation-role summary
method 2
citation-polarity summary
roles
method 2polarities
use method 2representative citing papers
DeepParse mines reusable regex patterns with an LLM from few log samples and applies them via Drain to achieve 97.6% average parsing accuracy on 16 datasets, outperforming baselines and cutting anomaly detection false alarms by over 30%.
RASLIK uses randomized antipodal search on linearized influence kernels to achieve data Pareto improvement in LLM unlearning, outperforming baselines with sublinear complexity and double gains in quality and efficiency.
Diverse teacher-generated rationales improve MLLM visual persuasiveness prediction via supervised fine-tuning, while a new three-dimensional faithfulness framework shows that prediction accuracy alone does not ensure faithful reasoning and that decision sensitivity best matches human preferences.
Benign fine-tuning of foundation models induces large, heterogeneous, and often contradictory changes in safety metrics across general and domain-specific benchmarks.
UAF is the first unified audio front-end LLM that turns multiple front-end tasks into one sequence prediction model processing streaming audio chunks and reference prompts to output semantic and control tokens for full-duplex interaction.
Traj-CoA is a multi-agent LLM framework that sequentially processes noisy five-year EHR data via worker agents into EHRMem for manager-agent lung cancer risk prediction and outperforms four categories of baselines in zero-shot evaluation.
WhisperRT converts Whisper to a causal streaming ASR model via encoder causality, decoder synchronization on partial states, and fine-tuning, achieving better performance than non-fine-tuned streaming methods on sub-300ms chunks with lower complexity.
Slot-MLLM introduces a slot-attention-based object-centric visual tokenizer with Q-Former encoder, diffusion decoder, and residual vector quantization for improved local visual comprehension and generation in multimodal LLMs.
Extending language model context length enables LMMs to process over 200K visual tokens from long videos without video training, achieving SOTA on Video-MME via dense frame sampling.
Zephyr-7B achieves state-of-the-art chat benchmark results among 7B models by distilling alignment via dDPO on AI feedback preferences, surpassing the 70B Llama-2-Chat model on MT-Bench with no human data required.
Introduces MAF framework and DeepModal-Bench to capture universal cross-modal forgery traces for better generalization in multimodal deepfake detection.
A small language model fine-tuned on tool-augmented chain-of-thought data generated by a larger LLM learns to selectively call tools, delivering better content moderation accuracy at lower inference cost.
InvDesFlow-AL combines active learning with diffusion generative models to improve crystal structure prediction accuracy by 33% and identifies Li2AuH6 as a candidate BCS superconductor with 140 K transition temperature.
Proposes and benchmarks a new aggregation technique for LoRA adapters in federated fine-tuning against existing methods on GLUE tasks.
A survey deriving a unified policy gradient framework for LLM post-training methods and providing technical comparisons of PPO, GRPO, DPO variants.
citing papers explorer
-
SteeringDiffusion: A Bottlenecked Activation Control Interface for Diffusion Models
SteeringDiffusion supplies a bottlenecked, prompt-conditioned activation interface for frozen diffusion models that delivers smooth monotonic content-style control via one runtime scalar and timestep gating.
-
DeepParse: Hybrid Log Parsing with LLM-Synthesized Regex Masks
DeepParse mines reusable regex patterns with an LLM from few log samples and applies them via Drain to achieve 97.6% average parsing accuracy on 16 datasets, outperforming baselines and cutting anomaly detection false alarms by over 30%.
-
Randomized Antipodal Search Done Right for Data Pareto Improvement of LLM Unlearning
RASLIK uses randomized antipodal search on linearized influence kernels to achieve data Pareto improvement in LLM unlearning, outperforming baselines with sublinear complexity and double gains in quality and efficiency.
-
Can MLLMs Reason About Visual Persuasion? Evaluating the Efficacy and Faithfulness of Reasoning
Diverse teacher-generated rationales improve MLLM visual persuasiveness prediction via supervised fine-tuning, while a new three-dimensional faithfulness framework shows that prediction accuracy alone does not ensure faithful reasoning and that decision sensitivity best matches human preferences.
-
Safety Drift After Fine-Tuning: Evidence from High-Stakes Domains
Benign fine-tuning of foundation models induces large, heterogeneous, and often contradictory changes in safety metrics across general and domain-specific benchmarks.
-
UAF: A Unified Audio Front-end LLM for Full-Duplex Speech Interaction
UAF is the first unified audio front-end LLM that turns multiple front-end tasks into one sequence prediction model processing streaming audio chunks and reference prompts to output semantic and control tokens for full-duplex interaction.
-
Traj-CoA: Patient Trajectory Modeling via Chain-of-Agents for Lung Cancer Risk Prediction
Traj-CoA is a multi-agent LLM framework that sequentially processes noisy five-year EHR data via worker agents into EHRMem for manager-agent lung cancer risk prediction and outperforms four categories of baselines in zero-shot evaluation.
-
WhisperRT -- Turning Whisper into a Causal Streaming Model
WhisperRT converts Whisper to a causal streaming ASR model via encoder causality, decoder synchronization on partial states, and fine-tuning, achieving better performance than non-fine-tuned streaming methods on sub-300ms chunks with lower complexity.
-
Slot-MLLM: Object-Centric Visual Tokenization for Multimodal LLM
Slot-MLLM introduces a slot-attention-based object-centric visual tokenizer with Q-Former encoder, diffusion decoder, and residual vector quantization for improved local visual comprehension and generation in multimodal LLMs.
-
Long Context Transfer from Language to Vision
Extending language model context length enables LMMs to process over 200K visual tokens from long videos without video training, achieving SOTA on Video-MME via dense frame sampling.
-
Zephyr: Direct Distillation of LM Alignment
Zephyr-7B achieves state-of-the-art chat benchmark results among 7B models by distilling alignment via dDPO on AI feedback preferences, surpassing the 70B Llama-2-Chat model on MT-Bench with no human data required.
-
Beyond Surface Artifacts: Capturing Shared Latent Forgery Knowledge Across Modalities
Introduces MAF framework and DeepModal-Bench to capture universal cross-modal forgery traces for better generalization in multimodal deepfake detection.
-
Tool-MCoT: Tool Augmented Multimodal Chain-of-Thought for Content Safety Moderation
A small language model fine-tuned on tool-augmented chain-of-thought data generated by a larger LLM learns to selectively call tools, delivering better content moderation accuracy at lower inference cost.
-
InvDesFlow-AL: active learning-based workflow for inverse design of functional materials
InvDesFlow-AL combines active learning with diffusion generative models to improve crystal structure prediction accuracy by 33% and identifies Li2AuH6 as a candidate BCS superconductor with 140 K transition temperature.
-
Aggregating Low Rank Adapters in Federated Fine-tuning
Proposes and benchmarks a new aggregation technique for LoRA adapters in federated fine-tuning against existing methods on GLUE tasks.
-
Reinforcement Learning for LLM Post-Training: A Survey
A survey deriving a unified policy gradient framework for LLM post-training methods and providing technical comparisons of PPO, GRPO, DPO variants.