Neural Collapse predicts that balanced one-hot classification pushes model representations to be equally far from each other; a symmetric configuration that depends only on the output label and ignores any semantic similarity in the inputs. This creates a puzzle: next-token prediction language models are trained predominantly (as context length increases) with one-hot labels: the same context is very unlikely to appear twice in training with different labels.
arXiv:2606. 26749v1 Announce Type: new Abstract: Neural Collapse predicts that balanced one-hot classification pushes model representations to be equally far from each other; a symmetric configuration that depends only on the output label and ignores any semantic similarity in the inputs.
arXiv:2608.30315v1 Announce Type: new Abstract: Token embeddings are the basic representational units that connect discrete tokens with continuous computation in language models. Although modern lang...
The paper investigates why token prediction, a common pre‑training objective for language models, yields useful representations. It introduces a statistical framework linking token prediction accuracy to the geometry of token embeddings, showing that accurate predictions organize embeddings according to Hellinger distances between context distributions. The authors also propose a self‑consistency principle that refines contextual representations through repeated application of a shared block, and provide downstream guarantees for token generation, community recovery, and linear classification.
arXiv:2606. 15521v1 Announce Type: cross Abstract: Tokenization introduces representational redundancy: under a fixed token vocabulary, every byte string admits many valid token encodings, or segmentations, that decode to the same surface string.
The paper investigates whether few-shot in-context learning (ICL) emerges similarly across different data modalities. Using a controlled cross-modality framework, the authors test the Convergent Emergence Hypothesis, which posits that tasks benefiting from ICL in one modality will also benefit in others. They find that paired-mapping ICL appears in six modalities—language, genome, integer sequences, time series, images, and proteins—outperforming baselines and showing correlated task effects in five of them, supporting the hypothesis in some but not all cases.
The paper examines how new vocabulary tokens are added to language models for generative recommendation tasks. It shows that the common practice of initializing these tokens as the mean of existing embeddings collapses them into a degenerate subspace, hindering fine‑tuning. The authors propose Grounded Token Initialization (GTI), which places new tokens at semantically meaningful positions in the pretrained embedding space using linguistic supervision, and demonstrate that GTI outperforms mean initialization and other adaptation methods across several benchmarks.
arXiv:2507. 01414v2 Announce Type: replace Abstract: We introduce a new family of toy problems that combine features of linear-regression-style continuous in-context learning (ICL) with discrete associative recall.
The paper investigates why large language models (LLMs) fail to maintain continuous mixtures of token embeddings—used in latent-state reasoning—to preserve multiple reasoning paths. Through theory and experiments, it identifies three failure sources: transformer geometry distortion, amplification or contraction dynamics from softmax and autoregressive feedback, and the need for context-dependent corrections that scale with mixture size. Empirical results confirm the predicted transition between contraction and amplification and show pretrained models largely fall on the amplifying side.
arXiv:2606. 07559v2 Announce Type: replace-cross Abstract: Fine-tuning a language model often fails silently when its correct completion must outrank a near-synonym competitor.
The paper compares latent representations in Selective State Space Models (SSMs) like Mamba and Transformers such as Pythia using Sparse Autoencoders. Across a 10‑million token corpus, 99.98% of Mamba features align closely with Pythia’s, supporting the Universality Hypothesis that core semantic representations are similar across architectures. A tiny 0.02% of features diverge, with Mamba’s recurrent bottleneck causing it to compress syntactic anomalies into polysemantic neurons, whereas Pythia’s attention can isolate distinct formatting edge‑cases.
The paper introduces TokenAdapt, a model‑agnostic tokenizer transplantation method that uses a hybrid heuristic to initialize new token embeddings, and a novel pre‑tokenization learning approach for multi‑word Supertokens to improve compression. TokenAdapt combines local subword decomposition and global semantic similarity to preserve semantics while reducing retraining needs. Empirical results show that TokenAdapt outperforms existing baselines such as Transtokenizer and ReTok, achieving lower perplexity ratios and significant compression gains.
The paper presents a mean‑field analysis of attention in language models, defining an average attention kernel that propagates representations layer by layer. When conditioned on a whole corpus, the kernel predicts the average evolution of representation geometry; when conditioned on a single context, it predicts the expected geometry for that context. The difference between actual attention and the mean‑field prediction—called the mean‑field deviation—captures context‑specific computation, revealing how models diverge from average behavior during training and in few‑shot tasks.