The paper introduces JoLT, a training‑free compressor that jointly allocates rank and precision for key‑value (KV) cache compression in long‑context language models. JoLT treats grouped prefill caches as fourth‑order tensors, applies partial Tucker decomposition along token and feature modes, and uses a rotated low‑bit quantizer for residuals, all governed by a single Lagrangian dual under a global byte constraint. Across five models from four architecture families, JoLT achieves 2–3× compression with less than 0.2% perplexity loss, and near‑lossless retrieval accuracy on LLaMA‑3.1‑8B at 64K context up to 3× compression.
By Rahul Krishnan, Volker Schulz
Lngram v2 introduces a latent N‑gram memory system that decouples memory routes, memory dimension, and backbone width, enabling scalable memory capacity for transformers. It employs context‑aware grouped‑query attention, a zero‑value sink, and counterfactual surrogate gradients to improve readout selectivity and routing trainability while preserving hard discrete addressing. Experiments on vision‑language models up to 30B parameters show consistent performance gains, reduced memory parameters, and stable semantic structure in the discrete IDs.
By Yunao Zheng, Bin Wen, Xiaojie Wang
arXiv:2608. 11661v1 Announce Type: cross Abstract: A multiplicative dual-encoder network computes a real-valued output for a pair of inputs as the inner product of their separate encodings.
By Zijian Zhao, Sen Li
hLLM (Hungarian LLM) is a decoding strategy that efficiently produces ranked lists from large language models by decoding all ordinal positions in a single forward pass. It extracts an item‑position score matrix from the model’s hidden states, then applies the Hungarian algorithm to obtain a valid permutation, avoiding the sequential token generation of traditional autoregressive decoding. Experiments show that LoRA‑based fine‑tuning with teacher ranking distillation achieves 28 ms end‑to‑end inference, a 64× speed‑up while preserving ranking quality comparable to the teacher model.
By Emil Laftchiev, Prachi Agrawal, Moe Kayali, Bixing Yan, Qi Xu, Zijie Lei, Chen Qiu, Zhi Hua, Ke Li, Luke Simon
arXiv:2601. 16991v3 Announce Type: replace-cross Abstract: Adapting large pre-trained language models to downstream tasks often entails fine-tuning millions of parameters or deploying costly dense weight updates, which hinders their use in resource-constrained environments.
By Longteng Zhang, Sen Wu, Shuai Hou, Zhengyu Qing, Zhuo Zheng, Danning Ke, Qihong Lin, Qiang Wang, Shaohuai Shi, Xiaowen Chu
The paper introduces the Von‑Neumann State‑Space Transformer (VN‑SST), a memory‑augmented Transformer that replaces the standard feed‑forward block with a low‑rank instruction bank. By decoding token‑specific operators from a low‑dimensional state‑space memory, VN‑SST achieves higher data‑efficiency and parameter‑efficiency on motor‑cortex neural‑decoding tasks and on small language‑model benchmarks. The model demonstrates that a compact instruction set can act as a control channel, improving performance without increasing accuracy through larger parameter counts.
By Morteza Sarafyazd
arXiv:2607. 15498v1 Announce Type: cross Abstract: The key-value (KV) cache is the main memory bottleneck in long-context large language model (LLM) inference.
By Shahrzad Esmat, Dhawal Shah, Ali Jannesari
The paper introduces SPD (Single Forward Pass), a decoding strategy that efficiently produces ranked lists from large language models by emitting only the ordinal positions of items. SPD extracts an item‑position score matrix from the model’s hidden states, then solves a bipartite assignment problem via the Hungarian algorithm to generate a valid permutation in a single forward pass. Experiments show that LoRA‑based fine‑tuning combined with autoregressive ranking distillation achieves 28 ms end‑to‑end inference, a 64× speed‑up while preserving ranking quality.
By Emil Laftchiev, Prachi Agrawal, Moe Kayali, Bixing Yan, Qi Xu, Zijie Lei, Chen Qiu, Zhi Hua, Ke Li, Luke Simon
arXiv:2606. 03879v1 Announce Type: cross Abstract: As foundation models scale toward fusing more heterogeneous visual streams, understanding how diverse encoders interact under joint training becomes a prerequisite for principled design.
By Wei Ding, Yudong Zhang, Ruobing Xie, Xingwu Sun, Jiansheng Chen, Yu Wang
arXiv:2606. 18811v1 Announce Type: cross Abstract: Learned sparse retrieval (LSR) models such as SPLADE have traditionally used BERT-style masked language models as backbone encoders.
By Youngjoon Jang, Seongtae Hong, Jonah Turner, Heuiseok Lim
NeoMME is a family of 260M and 800M‑parameter multimodal‑native multilingual encoders that process text and raw image patches in a single bidirectional Transformer. Trained from scratch with a masked discrete‑diffusion objective conditioned on visible image patches, NeoMME supports a 16,384‑token context, enabling encoding of up to two 4K UHD images. In downstream tests, NeoMME‑Retriever models outperform all sub‑800M‑parameter baselines on the ViDoRe v3 benchmark and achieve twice the throughput of ColModernVBERT on an NVIDIA L40S, while hierarchical token pooling and asymmetric quantization compress embeddings 255× with minimal loss in retrieval performance.
By Aur\'elien Lac, Tony Wu
Multimodal models often build on architectures designed for generative vision-language modeling, typically combining separately pretrained vision encoders with causal language models. Visual document...