REVIEW 14 cited by
Large Concept Models: Language Modeling in a Sentence Representation Space
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
LLMs have revolutionized the field of artificial intelligence and have emerged as the de-facto tool for many tasks. The current established technology of LLMs is to process input and generate output at the token level. This is in sharp contrast to humans who operate at multiple levels of abstraction, well beyond single words, to analyze information and to generate creative content. In this paper, we present an attempt at an architecture which operates on an explicit higher-level semantic representation, which we name a concept. Concepts are language- and modality-agnostic and represent a higher level idea or action in a flow. Hence, we build a "Large Concept Model". In this study, as proof of feasibility, we assume that a concept corresponds to a sentence, and use an existing sentence embedding space, SONAR, which supports up to 200 languages in both text and speech modalities. The Large Concept Model is trained to perform autoregressive sentence prediction in an embedding space. We explore multiple approaches, namely MSE regression, variants of diffusion-based generation, and models operating in a quantized SONAR space. These explorations are performed using 1.6B parameter models and training data in the order of 1.3T tokens. We then scale one architecture to a model size of 7B parameters and training data of about 2.7T tokens. We perform an experimental evaluation on several generative tasks, namely summarization and a new task of summary expansion. Finally, we show that our model exhibits impressive zero-shot generalization performance to many languages, outperforming existing LLMs of the same size. The training code of our models is freely available.
Forward citations
Cited by 14 Pith papers
-
Omnilingual SONAR: Cross-Lingual and Cross-Modal Sentence Embeddings Bridging Massively Multilingual Text and Speech
Progressive LLM-initialized training plus teacher-student distillation yields omnilingual sentence embeddings that halve FLORES error and cut BIBLE error 15× while supporting speech and zero-shot Spectrum transfer.
-
Hierarchical Latent Prediction for Language Models
HiLP adds a hierarchical latent prediction objective to LM pretraining, improving coding and multi-step reasoning benchmarks and speculative decoding acceptance, with zero inference-time overhead.
-
From Found to Designed: Concepts as a Design Axis for Large Language Models
Concepts should be an explicit design axis for LLMs, organized by pipeline stage and internal-vs-external origin, rather than recovered post-hoc.
-
From Interface to Inference: Eliciting Any-Order Inference from Any-Order Models
Masked diffusion language models fail to exploit their any-order interface because of positional uncertainty; insertion-based (FlexMDM) and latent-segment (LatentMDM) variants recover distinct any-order inference beha...
-
Concept Tokens: Learning Behavioral Embeddings Through Concept Definitions
A single token embedding trained on concept definitions can directionally steer a frozen LLM's behavior, but hallucination gains come mainly from abstention and new factual details are not reliably stored.
-
Controllably Efficient Language Models
A single transformer variant can compress past context into chunk summaries and use chunk size as a test-time knob to trade quality against speed and memory, outperforming many efficient baselines on recall benchmarks.
-
Mitigating Resolution-Drift in Federated Learning: Case of Keypoint Detection
RAF, resolution-adaptive federated learning, uses multi-resolution knowledge distillation to stop resolution differences between clients from degrading federated keypoint detection.
-
LLM Pretraining with Continuous Concepts
A language model trained to predict and interleave teacher-derived SAE concepts into its hidden states beats plain next-token prediction and knowledge distillation on several benchmarks.
-
An Annotated Reading of 'The Singer of Tales' in the LLM Era
LLM generation resembles oral-formulaic composition: single-pass, pattern-based, and non-authorial, so AI output should be treated as a new post-literate medium.
-
Synergy: End-to-end Concept Model
A byte-level transformer with a learned top-k router matches a tokenized Llama3 baseline on Wikipedia bits-per-byte, and works best when positional encoding is removed from its middle layers.
-
Improving Large Language Models with Concept-Aware Fine-Tuning
Adding lightweight multi-token auxiliary heads with a weighted future-token loss improves supervised fine-tuning of Llama-3-8B-Instruct across five diverse tasks.
-
TALL -- A Trainable Architecture for Enhancing LLM Performance in Low-Resource Languages
A trainable pipeline of translation models and a frozen LLM improves Hebrew last-word prediction accuracy to 5.59%, about twice the best baseline.
-
Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space
A training-free method that feeds probability-weighted token embeddings back into LLMs during reasoning, improving accuracy and token efficiency on math and coding tasks.
-
Concept-Level AI for Telecom: Moving Beyond Large Language Models
A position paper proposing Large Concept Models as the successor to LLMs for telecom network management, without experimental evidence.
Discussion (0). Continue with ORCID to comment.