The paper introduces the Proximity Trap, a problem where long-context language models fail to attend to distant evidence because of competition from abundant, irrelevant proximal background. To mitigate this, the authors propose LYRA, a t‑distributed directional matching mechanism that reshapes the retrieval distribution to focus more on task‑relevant evidence while preserving positional information. Experiments on LongBench‑v2, RULER, and LongBench show consistent performance gains, and a new benchmark, ProxBench, is presented to evaluate distant evidence usage under varying proximal interference.
arXiv:2606. 29844v1 Announce Type: cross Abstract: The quadratic computational cost of traditional attention mechanisms poses a major bottleneck to the scalability and practical deployment of large language models (LLMs), particularly in long-context scenarios.
arXiv:2609.38530v1 Announce Type: new Abstract: Language models increasingly use architectures that vary attention span and positional encoding across layers, such as applying RoPE with sliding-windo...
arXiv:2511. 10696v3 Announce Type: replace-cross Abstract: Sparse attention is crucial in long-context Transformers, which restricts each token to a limited neighborhood and thereby reduces the quadratic cost of full self-attention.
The paper introduces GAPE (Gated Adaptive Positional Encoding), a drop‑in augmentation for rotary positional encodings that adds content‑aware bias to attention logits while preserving rotary geometry. GAPE uses a query‑dependent gate to contract irrelevant context and a key‑dependent gate to keep salient distant tokens, leading to exponential attenuation of weakly protected context. Experiments show that GAPE improves long‑context robustness, extrapolating up to eight times the training length, and can be retrofitted into a pretrained 7B model without losing benchmark performance.
arXiv:2610.08463v1 Announce Type: cross Abstract: Long-context inference and Retrieval-Augmented Generation (RAG) handle evidence selection at vastly different scales, from a single long prompt to an...
Long-context inference and Retrieval-Augmented Generation (RAG) handle evidence selection at vastly different scales, from a single long prompt to an entire corpus. We ask whether a single model-inter...
arXiv:2609.22101v1 Announce Type: cross Abstract: Large language models can process increasingly long prompts, yet their ability to locate and use decisive evidence may degrade as irrelevant or confu...
The paper investigates how the dimension of attention heads influences long‑context retrieval in language models. Experiments show that while pretraining loss and short‑context tasks are largely unaffected, wider heads significantly improve retrieval performance at fixed model sizes. Ablation studies indicate that this benefit persists even when attention weights are frozen, suggesting that wider heads enhance adaptation of other model components such as embeddings and MLPs.
The paper introduces EXACT, a supervision-allocation objective that reweights long-context targets by inverse frequency to address a token-level supervision mismatch in packed training. Across seven Qwen/LLaMA configurations, EXACT consistently improves NoLiMa and RULER metrics, with notable gains on Qwen2.5-0.5B and LLaMA-3.2-3B. Standard QA and reasoning performance remain largely unchanged, and a distance-resolved probe shows benefits primarily when evidence is thousands of tokens away.
arXiv:2607. 06160v1 Announce Type: cross Abstract: Synthesizing long-context supervised fine-tuning (SFT) data is a scalable way to enhance the long-context understanding of large language models (LLMs), yet existing approaches share three limitations: narrow task coverage, insufficient instruction difficulty, and a lack of faithfulness supervision.
arXiv:2608. 03297v1 Announce Type: new Abstract: A standard claim in the literature on retrieval-augmented and memory-augmented language models is that shorter context is better when the relevant information is preserved.
arXiv:2607. 02980v1 Announce Type: cross Abstract: Scaling modern large language models (LLMs) to long contexts is limited by the quadratic computation cost, and poor length extrapolation of dense attention.