arXiv AI
Jun 15
Planning with the Views via Scene Self-Exploration
arXiv:2605. 29563v2 Announce Type: replace Abstract: Can VLMs predict how each camera move changes the view, and plan many such moves ahead?
Inference efficiency
Quantization, distillation, pruning and serving work aimed at the same accuracy for less memory, latency and money.
arXiv AI
Jun 15
Minim: Privacy-Aware Minimal View for Agents via Trusted Local Sanitization
arXiv:2606. 13949v1 Announce Type: new Abstract: Modern LLM-powered autonomous agents increasingly rely on rich user interface (UI) state observations to achieve reliable action grounding in complex digital environments.
arXiv AI
Jun 15
When Sample Selection Bias Precipitates Model Collapse
arXiv:2606. 13732v1 Announce Type: new Abstract: The proliferation of recursive training on synthetic data can alleviate data scarcity but risks model collapse, where repeated training erodes distributional tails and homogenizes outputs.
arXiv Machine Learning
Jun 15
Efficient Rationale-based Retrieval: On-policy Distillation from Generative Rerankers based on JEPA
arXiv:2604. 23336v3 Announce Type: replace-cross Abstract: Unlike traditional fact-based retrieval, rationale-based retrieval typically necessitates cross-encoding of query-document pairs using large language models, incurring substantial computational costs.
arXiv Machine Learning
Jun 15
Be My Tutor: On-Policy Co-Distillation for Mutual LLM Improvement via Peer Feedback
arXiv:2606. 14368v1 Announce Type: new Abstract: We study multi-domain LLM training in which two models, each stronger in a different domain, co-evolve by tutoring each other through on-policy feedback.
arXiv Machine Learning
Jun 15
Decompose Sparsely Where You Should, Absorb Densely Where You Should No
arXiv:2606. 14040v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) are typically trained to reconstruct the \textbf{entire} residual stream through a sparse dictionary, implicitly assuming that all activation content is amenable to sparse, monosemantic decomposition.
arXiv AI
Jun 15
Quantized Evolution Strategies: High-precision Fine-tuning of Quantized LLMs at Low-precision Cost
arXiv:2602. 03120v2 Announce Type: replace-cross Abstract: Post-Training Quantization (PTQ) is essential for deploying Large Language Models (LLMs) on memory-constrained devices, yet it renders models static and difficult to fine-tune.
arXiv AI
Jun 15
Listening with Attention: Entropy-Guided Explainability for Transformer-Based Audio Models
arXiv:2606. 14647v1 Announce Type: cross Abstract: Transformer-based automatic speech recognition (ASR) models such as Whisper are highly accurate, but their predictions remain difficult to interpret.
arXiv AI
Jun 15
Moonlight in Latent Space: Chirality and Structural Correspondence Between Beethoven's Op. 27 No. 2 and Machine Learning Mechanisms
arXiv:2606. 14612v1 Announce Type: cross Abstract: We show that the three movements of Beethoven's "Moonlight Sonata" (Op.
arXiv AI
Jun 15
Simplex-Constrained Sparse Bagging: Transitioning from Uniform Priors to Sparse Posteriors in Ensemble Learning
arXiv:2606. 13589v1 Announce Type: cross Abstract: We present Simplex-Constrained Sparse Bagging (SCSB), a mathematically rigorous framework for post-training compression and probability calibration of bootstrap-based bagging ensembles.
arXiv AI
Jun 15
CARE: Controlling LLM-Generated Policies through Auditable Review of Evidence in Scientific Experimentation
arXiv:2606. 14581v1 Announce Type: cross Abstract: Granting LLMs direct control over costly, irreversible scientific experiments leads to unsafe exploration and unstable performance, but discarding LLM creativity entirely sacrifices significant optimization potential.
arXiv Machine Learning
Jun 15
Generative Modeling of Bach-Style Symbolic Music: A Comparative Study of Autoregressive, Latent-Variable, and Adversarial Approaches
arXiv:2606. 13626v2 Announce Type: replace-cross Abstract: We study generative modeling of Bach-style symbolic piano music using a shared MIDI corpus and three model families: autoregressive LSTMs with attention, latent-variable models including recurrent VAEs and vector-quantized VAEs, and generative adversarial networks.
arXiv Machine Learning
Jun 15
Sub-Token Routing for KV Cache Compression
arXiv:2604. 21335v3 Announce Type: replace Abstract: Transformer inference often requires a large KV cache, especially for long-context language modeling and multimodal generation.
arXiv AI
Jun 15
Beyond LoRA: Is Sparsity-Induced Adaptation Better?
arXiv:2606. 13767v1 Announce Type: cross Abstract: Low-rank adaptation (LoRA) and its variants provide a memory- and compute-efficient alternative to full fine-tuning of pre-trained models.
arXiv AI
Jun 15
Shift-Invariant Attribute Scoring for Kolmogorov-Arnold Networks via Shapley Value
arXiv:2510. 01663v2 Announce Type: replace-cross Abstract: For many real-world applications, understanding feature-outcome relationships is as crucial as achieving high predictive accuracy.
arXiv Machine Learning
Jun 15
Realizing Native INT8 Compute for Diffusion Transformers on Consumer GPUs: A Fused INT8 GEMM Kernel for Ideogram 4.0
arXiv:2606. 14598v1 Announce Type: new Abstract: Post-training INT8 (W8A8) quantization of diffusion transformers is widely deployed as a speed optimization, yet on consumer Ampere GPUs it is frequently slower than the FP8 and NF4 alternatives it is meant to beat.
arXiv AI
Jun 15
Squeeze-Release: Iterative Pruning with Exact Structural Minimization
arXiv:2606. 14346v1 Announce Type: cross Abstract: Unstructured pruning produces sparse weight tensors, but the standard implementation keeps tensor shapes unchanged so the deployed model is no smaller than before pruning.
arXiv AI
Jun 15
Fragile Knowledge, Robust Instruction-Following: The Width Pruning Dichotomy in Llama-3.2
arXiv:2512. 22671v3 Announce Type: replace-cross Abstract: Structured width pruning of GLU-MLP layers in Llama-3.
arXiv AI
Jun 15
Rethinking the Trust Region in LLM Reinforcement Learning
arXiv:2602. 04879v3 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has become a cornerstone for fine-tuning Large Language Models (LLMs), with Proximal Policy Optimization (PPO) serving as the de facto standard algorithm.
arXiv Machine Learning
Jun 15
HumP-KD: A Hybrid Uncertainty-Aware Multi-Stage Progressive Knowledge Distillation Framework for Efficient Fire Classification
arXiv:2606. 14684v1 Announce Type: cross Abstract: Real-time fire classification systems require models that are simultaneously accurate, computationally efficient, and deployable on resource-constrained hardware.