Cosine similarity / Cosine similarity analysis / Cosine semantic similarity / cosine similarity of hidden states / Sample-wise cosine similarity / Cosine similarity of attention maps / Cosine similarity perturbation analysis / Cosine similarity template matching / Cosine similarity to neighbours / Semantic consistency (cosine similarity)

anchor

Normalised dot product between two vectors. It measures directional alignment independently of vector magnitude; in the TerraMind concept-vector report it evaluates alignment with the empirical embedding shift.

Findings

• IC-005GPT-4o underperforms AHA and other VLMs in detecting and reasoning about robotic manipulation failures across multiple datasets. [eval]
• IC-022Safety layers in Llama-2-chat-7b show substantial divergence during VL adaptation, correlating with safety degradation [primary]
• IC-1201CLIP-ViT-L/14 image features support 200-way zero-shot EEG-based object recognition better than ViT-B/16 or ResNet-50 features when used as a frozen encoder in a contrastive learning framework [eval]
• IC-1458GPT-4 and PALM 2-L produce significantly less consistent descriptions of interpolated domain embeddings than a purpose-built ELM model [eval]
• IC-148Language models represent semantically equivalent inputs from different data types (languages, code, images, audio) close together in intermediate layers, with the shared space scaffolded by the model's dominant language [primary]
• IC-480Pre-trained M-LLMs (GPT-4o, LLaVA-v1.6-34B, InternVL2-26B, Qwen2-VL-7B) produce imprecise tampering explanations when artifacts require fine-grained pixel-level analysis such as lighting or perspective inconsistencies [eval]
• IC-548CLIP-B/32 exhibits progressively increasing layer-wise representation similarity in both its vision encoder and text encoder, and the pattern also holds across modalities [primary]
• IC-565Phi-3's residual stream encodes format instructions as linear directions, evidenced by cosine similarity and vocabulary-space projections [primary]
• IC-579ICL hidden states exhibit positional bias: representations of the same input are more similar when the input appears at similar positions in the sequence [primary]
• IC-646DINOv2, DeiT-III, and OpenCLIP repurpose approximately 2% of patch tokens in low-informative background areas as internal registers, discarding local patch information while aggregating global image information; DINO does not exhibit this behaviour [primary]
• IC-914In Pythia models (70M through 2.8B), BERT-base, OPT-6.7B, LLaMA-2-7B, and ViT-Huge, the MLP out-projection vectors are almost orthogonal throughout training [eval]
• TM-002The mean-difference concept vector scores higher than trained classifiers under cosine similarity [eval]