In-weights learning induces linear embeddings enabling transitive inference in transformers, whereas in-context learning defaults to match-and-copy unless pre-trained on linear tasks or prompted with linear mental maps.
hub
Do large language models latently perform multi-hop reasoning? arXiv preprint arXiv:2402.16837
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Coconut lets LLMs perform reasoning directly in continuous latent space by recycling hidden states as inputs, outperforming standard chain-of-thought on search-intensive logical tasks with better accuracy-efficiency trade-offs.
Power-law data sampling creates beneficial asymmetry in the loss landscape that lets models acquire high-frequency skill compositions first, enabling more efficient learning of rare long-tail skills than uniform distributions.
HypEHR is a hyperbolic embedding model for EHR data that uses Lorentzian geometry and hierarchy-aware pretraining to answer clinical questions nearly as well as large language models but with much smaller size.
EHRAG constructs structural hyperedges from sentence co-occurrence and semantic hyperedges from entity embedding clusters, then applies hybrid diffusion plus topic-aware PPR to retrieve top-k documents, outperforming baselines on four datasets with linear indexing cost and zero token overhead.
SeLaR selectively applies latent soft reasoning in LLMs via entropy gating and contrastive regularization, outperforming standard CoT on five benchmarks without training.
SCoRe uses multi-turn online RL with regularization on self-generated traces to improve LLM self-correction, achieving 15.6% and 9.1% gains on MATH and HumanEval for Gemini models.
DiscoLoop adds a discrete embedding channel to looped transformers to fix representational misalignment in two-hop reasoning, yielding near-perfect accuracy on synthetic tasks and better pretraining loss on real data.
Frontier AI models' no-CoT 50% task-completion time horizons have doubled yearly over six years, reaching over 3 minutes for GPT-5.5 with projections to 25 minutes by 2030.
Injecting noise into LLM latent trajectories creates diverse reasoning paths whose agreement acts as a confidence signal for selective abstention, cutting error rates from 40-70% to under 15% on math tasks.
LACE enables concurrent reasoning paths in LLMs to interact via lattice attention and a synthetic training pipeline, raising accuracy more than 7 points over independent parallel search.
Heima compresses verbose CoT into hidden thinking tokens via information-theoretic analysis and an adaptive interpreter, claiming maintained or improved zero-shot accuracy on reasoning benchmarks.
citing papers explorer
-
LACE: Lattice Attention for Cross-thread Exploration
LACE enables concurrent reasoning paths in LLMs to interact via lattice attention and a synthetic training pipeline, raising accuracy more than 7 points over independent parallel search.