paper_id	claim	claim_status	evidence_status	source_depth	source_quote	page_or_section	taxonomy_fit	audit_status
2507.23083v1	In this work, we propose CARoPE (ContextAware Rotary Positional Embedding), a novel generalization of RoPE that dynamically generates head-specific frequency patterns conditioned on token embeddings.	supported	has evidence row	full-text	For example, at a se- quence length of 1024, CARoPE reduces perplex- ity by more than 60% compared to RoPE in the GPT-Tiny model (36.74 vs. 81.27).	p1	in-scope: taxonomy category match	pass; full-text verified; report=audit_report.md
2511.09146v2	•We show that RoPE’slow-frequency alignmentinduces attention heads with long-range dependency capability, while extrapolative heads are intrinsically low-rank and benefit from preserved positional encoding.	supported	has evidence row	full-text	Among many ap- proaches (Press et al., 2021; Chen et al., 2023b; Su et al., 2024; Peng et al., 2023; Wang et al., 2021), Rotary Position Embedding (RoPE) (Su et al., 2024) is widely used because it encodes rel- ative positions within dot-product attention and often extrapolates well to longer contexts.	p1	in-scope: taxonomy category match	pass; full-text verified; report=audit_report.md
2104.09864v5	We introduce a novel method, namely Rotary Position Embedding(RoPE), to leverage the positional information into the learning process of PLMS.	supported	has evidence row	full-text	However, when increasing the maximum input text length to 1024, RoFormer outperforms WoBERT by an absolute improvement of 1.5%.	p2	in-scope: taxonomy category match	pass; full-text verified; report=audit_report.md
2604.09742v1	While the rotation in RoPE can be efficiently implemented using matrix operations, the accompanying split and merge steps—implemented as vector operations—introduce non-negligible computational overhead.	supported	has evidence row	full-text	Metrics.We report bothspeedup timest 0/tand speedup percentage(t 0−t)/t 0, wheret 0denotes the baseline runtime andtdenotes the optimized runtime using RoME. 5.2.	p2	in-scope: taxonomy category match	pass; full-text verified; report=audit_report.md
2502.11664v4	To overcome these issues, we propose Video Rotary Position Embedding (VRoPE), a novel positional encoding method tailored for Video-LLMs.	supported	has evidence row	full-text	Specifically, VRoPE achieves an accuracy that is 32.19 points higher than RoPE and 14.22 points higher than RoPE-3D when the number of input frames increases to 1024-1216.	p1	in-scope: taxonomy category match	pass; full-text verified; report=audit_report.md