Quantization, distillation, pruning and serving work aimed at the same accuracy for less memory, latency and money.
arXiv:2606. 11576v1 Announce Type: cross Abstract: Modern Vision-Language Models (VLMs) benefit from chain-of-thought prompting and test-time scaling, but these gains often come with prohibitive inference cost due to large visual contexts and long decoding chains.
arXiv:2602. 10908v2 Announce Type: replace-cross Abstract: We present SoftMatcha 2, an ultra-fast and flexible search algorithm that enables search over trillion-scale natural language corpora in under 0.
arXiv:2605. 29128v2 Announce Type: replace Abstract: The wide adoption of LLMs has led to their use in great variety of applications and scenarios, such as chatbot assistants and data annotation, creating the need for the models to satisfy certain budget and hardware constraints.
arXiv:2506. 20040v3 Announce Type: replace-cross Abstract: Interpreting language models remains challenging due to the existence of residual stream, which linearly mixes and duplicates features across adjacent layers, causing single-layer analyses to miss this cross-layer structure.
arXiv:2606. 12412v1 Announce Type: cross Abstract: Vision-language models (VLMs) project images into hundreds to thousands of visual tokens, making decoder inference expensive in both attention computation and KV-cache memory.
arXiv:2606. 11391v1 Announce Type: new Abstract: Tensor Product Representations provide the structural fidelity required for symbolic reasoning in models but suffer from exponential dimensionality growth when encoding deep recursive structures.
arXiv:2606. 11556v1 Announce Type: cross Abstract: Continuous electrocardiography (ECG) monitoring could surface rhythm abnormalities before they escalate into cardiovascular events.
arXiv:2606. 12278v1 Announce Type: cross Abstract: Neural network pruning reduces model size by removing less important parameters while aiming to preserve predictive performance.
arXiv:2606. 11709v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) provides dense, token-level supervision for reasoning models by aligning a model's own distribution with the distribution it produces under privileged context, typically a verified solution.
arXiv:2606. 11761v1 Announce Type: new Abstract: Dynamic data pruning techniques aim to reduce computational cost while minimizing information loss by periodically selecting representative subsets of input data during model training.
arXiv:2601. 17360v2 Announce Type: replace-cross Abstract: An adversary observing a model's released prediction can infer sensitive attributes of the queried input, or even reconstruct representatives of the model's training data.
arXiv:2606. 11854v1 Announce Type: cross Abstract: There are two main Parameter-Efficient Fine-Tuning (PEFT) techniques for Large Language Models (LLMs).
arXiv:2606. 11844v1 Announce Type: new Abstract: Continual anomaly detection in tabular data is challenging and remains largely underexplored, particularly in settings with heterogeneous feature schemas, distribution shifts, and severe class imbalance.
Regression trees are among the most interpretable yet expressive model classes in machine learning. Historically, greedy induction has been the dominant approach for constructing well-performing regression trees.
Modern conversational agents condition on an ever-growing dialogue history at each turn, incurring redundant attention and encoding costs that grow with conversation length. Naive truncation or summarization degrades fidelity, while existing context compressors lack cross-turn memory sharing or revision, causing information loss and compounding errors in long dialogues.
Reinforcement learning (RL) has become a key component in modern large language models, yet the rollout stage remains the key bottleneck in RL training pipelines. Although Multi-Token Prediction (MTP) offers a natural solution to accelerate rollouts through speculative decoding, many studies have observed that MTP acceptance rates degrade significantly during RL training, leading to limited speedup performance.
Neural network pruning reduces model size by removing less important parameters while aiming to preserve predictive performance. Although the Lottery Ticket Hypothesis (LTH) shows that sparse subnetworks can match dense networks when trained from suitable initializations, its iterative pruning procedure requires multiple complete training cycles.
We propose a multi-agent collaborative framework built upon a lightweight Multimodal Large Language Model (MLLM), specifically designed for social intelligence reasoning. A key feature of our approach is that both the training and inference phases are augmented via knowledge distillation.
On-policy self-distillation (OPSD) provides dense, token-level supervision for reasoning models by aligning a model's own distribution with the distribution it produces under privileged context, typically a verified solution. However, we show that the learning signal drawn from this distributional gap concentrates on style tokens rather than task-bearing ones, as the hinted model tends to produce more direct, shorter outputs.
Deep search requires agents to answer complex questions through multi-step web search, browsing, evidence comparison, and synthesis. A central challenge is deciding how to search when several directions look plausible but only some will later lead to reliable evidence.