arXiv:2610. 10552v1 Announce Type: cross Abstract: Comparing parameter-efficient fine-tuning recipes under a single, shared learning rate is a common but flawed practice: when the arms being compared have very different trainable-parameter counts, a shared rate can simultaneously depress the larger arms' means and inflate their variance, manufacturing a large, seemingly multi-seed-significant advantage for the smallest arm that is not a real effect.
arXiv:2608. 01247v1 Announce Type: cross Abstract: Query-agnostic KV cache eviction compresses a context once and reuses the resulting cache for arbitrary future queries, but performance can collapse under tight budgets.
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.
arXiv:2609.37988v1 Announce Type: new Abstract: As the context size of text processed with an LLM grows, the size of KV caches can outstrip the memory allocated for the original model weights. This i...
arXiv:2607. 15456v1 Announce Type: new Abstract: Looped, weight-tied Transformers reduce parameters by reusing a block, but decoding still stores a separate K/V cache for every recurrence step.
Query-agnostic KV cache eviction compresses a context once and reuses the resulting cache for arbitrary future queries, but performance can collapse under tight budgets. Existing methods primarily improve which original KV pairs are retained.
arXiv:2609.17109v1 Announce Type: new Abstract: A common small-model deployment runs one shared backbone with several LoRA specialists that answer over the same context. Serving them naively re-prefi...
The paper introduces Draft‑Guided Eviction (DGE), a training‑free KV‑cache compression technique that postpones token eviction until after drafting the first two answer tokens, thereby preserving the cache entries needed for the answer trajectory. Unlike prior methods that focus on optimizing what to keep, DGE changes the timing of eviction, allowing existing compression schemes such as SnapKV, H2O, and PyramidKV to be applied without altering their eviction scores. Experiments show DGE consistently outperforms previous approaches across multiple instruct‑tuned models, achieving near‑full‑cache performance on LongBench and demonstrating that the improvement stems from trajectory anchoring rather than changes in cache selection.
arXiv:2608. 11231v1 Announce Type: new Abstract: LLM serving is increasingly accelerated by position-independent caching (PIC).
arXiv:2610.02953v1 Announce Type: new Abstract: Long-context LLM serving is increasingly bottlenecked by KV-cache memory, especially in resource-constrained scenarios. Among existing KV-cache compres...
The paper introduces iS-KV, an online low‑rank KV‑cache compression technique that uses block‑incremental SVD to manage memory during long‑horizon autoregressive decoding. Unlike token‑eviction methods, iS‑KV retains all positions in a compact representation by keeping a recent window exact and incrementally folding older states into bounded‑rank bases, synchronizing coordinates as the basis evolves. Experiments on DeepSeek‑R1‑Distill‑Llama‑8B and Qwen3‑8B show that iS‑KV achieves high accuracy (82.6% and 89.2% respectively) while providing 4.06‑fold and 5.64‑fold compression, outperforming token‑eviction baselines under matched memory budgets.
A common small-model deployment runs one shared backbone with several LoRA specialists that answer over the same context. Serving them naively re-prefills that shared context once per specialist. We s...
AbridgeKV compresses long chain‑of‑thought outputs by encoding completed text blocks into the KV cache of specially trained pause tokens, freeing the original cache space. The method uses pause tokens positioned within the block, allocates cache budget per layer based on entry value, and compresses large blocks. Experiments on 446 competition problems show that with only a quarter of the original KV cache, AbridgeKV matches the accuracy and output length of a full‑KV model, while enabling faster inference and a 3.5× smaller cache per sequence.