Search papers, labs, and topics across Lattice.
This paper introduces Right Reset (RR), a novel probing method that leverages prefix removal to identify effective boundaries in causal language models, allowing for the recovery of coherent text chunks. By applying RR to flattened text from topically similar records, the authors achieve a significant improvement in chunk recovery, with 47.7% of original records recovered compared to 25.9% using the strongest conventional baseline. The findings highlight that context dependence can serve as a valuable signal for boundary detection, even in the absence of clear surface structures.
Recovering nearly half of the original text records as clean units reveals that context dependence can effectively signal boundaries in language models, outperforming conventional methods.
Removing the left context from a causal language model reveals a useful kind of boundary: an edge where the model processes the same right-hand tokens with little change. We turn this observation into prefix-removal probing and introduce Right Reset (RR), which measures preservation of the right-hand hidden-state trajectory. A dynamic program converts RR edge scores into variable-length chunks. On flattened text formed by concatenating topically similar records after deleting their separators and layout, RR recovers 47.7% of the original records as clean units, versus 25.9% for a BGE embedding boundary, the strongest tested conventional baseline without task-specific model training. The gain persists after rendering and OCR. Passive scores from the same Qwen3-4B layer and direct prompting of a same-scale instruction model perform substantially worse on flattened records. Across six language models, RR-selected cuts also undergo consistently less local output disruption than unselected candidate edges. An observed-token likelihood-ratio readout is competitive in some architectures, indicating that the central contribution is the intervention: context dependence itself can provide a boundary signal when surface structure is weak.
