Pith. sign in

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

arxiv 2412.08821 v2 pith:46HOFOF2 submitted 2024-12-11 cs.CL

classification cs.CL
keywords conceptmodelmodelssentencespacelargellmstraining
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
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.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 14 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Omnilingual SONAR: Cross-Lingual and Cross-Modal Sentence Embeddings Bridging Massively Multilingual Text and Speech

    cs.CL 2026-03 accept novelty 7.0 of 10

    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.

  2. Hierarchical Latent Prediction for Language Models

    cs.CL 2026-08 conditional novelty 6.0 of 10

    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.

  3. From Found to Designed: Concepts as a Design Axis for Large Language Models

    cs.CL 2026-07 conditional novelty 6.0 of 10

    Concepts should be an explicit design axis for LLMs, organized by pipeline stage and internal-vs-external origin, rather than recovered post-hoc.

  4. From Interface to Inference: Eliciting Any-Order Inference from Any-Order Models

    cs.LG 2026-07 conditional novelty 6.0 of 10

    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...

  5. Concept Tokens: Learning Behavioral Embeddings Through Concept Definitions

    cs.CL 2026-01 conditional novelty 6.0 of 10

    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.

  6. Controllably Efficient Language Models

    cs.LG 2025-11 conditional novelty 6.0 of 10

    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.

  7. Mitigating Resolution-Drift in Federated Learning: Case of Keypoint Detection

    cs.CV 2025-07 conditional novelty 6.0 of 10

    RAF, resolution-adaptive federated learning, uses multi-resolution knowledge distillation to stop resolution differences between clients from degrading federated keypoint detection.

  8. LLM Pretraining with Continuous Concepts

    cs.LG 2025-02 conditional novelty 6.0 of 10

    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.

  9. An Annotated Reading of 'The Singer of Tales' in the LLM Era

    cs.CY 2025-02 conditional novelty 6.0 of 10

    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.

  10. Synergy: End-to-end Concept Model

    cs.CL 2025-07 reject novelty 5.0 of 10

    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.

  11. Improving Large Language Models with Concept-Aware Fine-Tuning

    cs.LG 2025-06 conditional novelty 5.0 of 10

    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.

  12. TALL -- A Trainable Architecture for Enhancing LLM Performance in Low-Resource Languages

    cs.CL 2025-06 reject novelty 5.0 of 10

    A trainable pipeline of translation models and a frozen LLM improves Hebrew last-word prediction accuracy to 5.59%, about twice the best baseline.

  13. Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space

    cs.CL 2025-05 conditional novelty 5.0 of 10

    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.

  14. Concept-Level AI for Telecom: Moving Beyond Large Language Models

    cs.NI 2025-06 reject novelty 4.0 of 10

    A position paper proposing Large Concept Models as the successor to LLMs for telecom network management, without experimental evidence.

Pith tools