Looped GPT-BERT: Trading Parameters for Computation in Small Language Modeling

• URL: http://arxiv.org/abs/2609.09691v1
• Date: Wed, 09 Sep 2026 04:14:23 GMT
• Title: Looped GPT-BERT: Trading Parameters for Computation in Small Language Modeling
• Abstract summary: We study Looped GPT-BERT in the BabyLM 2026 Strict-small setting, combining GPT-BERT's masked next-token and causal language-modeling objectives with depth-wise parameter sharing.<n>We train on a pre 7.48M-word English corpus and compare objective ratios, non-looped and looped architectures, and loop counts.<n>The BabyLM 2026 leaderboard reports an Overall Average of 35.42 and an NLP Average of 48.48.
• Score: 6.938321637054074
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: When training data are limited, increasing parameter count is not the only way to improve language-model performance. A small parameter set, when repeatedly applied, can also deliver comparable performance. We study Looped GPT-BERT in the BabyLM 2026 Strict-small setting, combining GPT-BERT's masked next-token and causal language-modeling objectives with depth-wise parameter sharing. We train on a preprocessed 7.48M-word English corpus and compare objective ratios, non-looped and looped architectures, and loop counts. Our final $4\times12$ model uses four physical layers for twelve recurrent traversals and contains 12.18M parameters. The BabyLM 2026 leaderboard reports an Overall Average of 35.42 and an NLP Average of 48.48. Compared with public BabyLM 10M Strict-small GPT-2 and GPT-BERT baselines, it achieves comparable performance on selected linguistic and downstream metrics, including BLiMP and GLUE, with fewer parameters. The loop ablations show that additional recurrent computation can improve training and preserve strong performance on selected linguistic tasks, whereas poorer performance on other tasks may reveal an inherent limitation of the looped design: using only a few physical layers restricts the model's representational space.

Related papers

• Efficient Multilingual Neural Machine Translation via Corpus-Driven Vocabulary Pruning: An English-Arabic Case Study [0.19921702137386235]
  We propose a general optimization framework that combines a vocabulary pruning method with a targeted fine-tuning protocol for MNMT models.<n>We evaluate the proposed framework using three models (M2M100, NLLB-200, mBART-50) on the English-Arabic language pair.<n>Results show that optimized multilingual models can match or exceed the performance of dedicated bilingual baselines.
  arXiv  Detail & Related papers  (2026-08-04T11:17:54Z)
• MMTEB: Massive Multilingual Text Embedding Benchmark [85.18187649328792]
  We introduce the Massive Multilingual Text Embedding Benchmark (MMTEB)<n>MMTEB covers over 500 quality-controlled evaluation tasks across 250+ languages.<n>We develop several highly multilingual benchmarks, which we use to evaluate a representative set of models.
  arXiv  Detail & Related papers  (2025-02-19T10:13:43Z)
• BESA: Pruning Large Language Models with Blockwise Parameter-Efficient Sparsity Allocation [54.28841287750586]
  Large language models (LLMs) have demonstrated outstanding performance in various tasks, such as text summarization, text question-answering, and etc. Existing solutions such as SparseGPT and Wanda attempt to alleviate this issue through weight pruning. This paper introduces a novel LLM pruning technique dubbed blockwise parameter-efficient sparsity allocation (BESA) by applying a blockwise reconstruction loss.
  arXiv  Detail & Related papers  (2024-02-18T12:44:15Z)
• Large Language Models as Zero-shot Dialogue State Tracker through Function Calling [42.00097476584174]
  We propose a novel approach for solving dialogue state tracking with large language models (LLMs) through function calling. This method improves zero-shot DST, allowing adaptation to diverse domains without extensive data collection or model tuning. We show that our approach achieves exceptional performance with both modestly sized open-source and also proprietary LLMs.
  arXiv  Detail & Related papers  (2024-02-16T06:13:18Z)
• Contrastive Preference Optimization: Pushing the Boundaries of LLM Performance in Machine Translation [50.00235162432848]
  We train ALMA models with only 22K parallel sentences and 12M parameters. The resulting model, called ALMA-R, can match or exceed the performance of the WMT competition winners and GPT-4.
  arXiv  Detail & Related papers  (2024-01-16T15:04:51Z)
• Text Alignment Is An Efficient Unified Model for Massive NLP Tasks [24.069447197357164]
  Next-word prediction is often not an efficient formulation for many NLP tasks. We propose text alignment as an efficient unified model for a wide range of crucial tasks. Our model delivers on par or even superior performance with much smaller model sizes.
  arXiv  Detail & Related papers  (2023-07-06T02:28:31Z)
• Large Language Models are Effective Text Rankers with Pairwise Ranking Prompting [65.00288634420812]
  Pairwise Ranking Prompting (PRP) is a technique to significantly reduce the burden on Large Language Models (LLMs) Our results are the first in the literature to achieve state-of-the-art ranking performance on standard benchmarks using moderate-sized open-sourced LLMs.
  arXiv  Detail & Related papers  (2023-06-30T11:32:25Z)
• Effects of sub-word segmentation on performance of transformer language models [0.628122931748758]
  We compare GPT and BERT models trained with the statistical segmentation algorithm BPE vs. two unsupervised algorithms for morphological segmentation. We show that training with morphological segmentation allows the LMs to: 1. achieve lower perplexity, 2. converge more efficiently in terms of training time, and 3. achieve equivalent or better evaluation scores on downstream tasks.
  arXiv  Detail & Related papers  (2023-05-09T14:30:29Z)
• Massively Multilingual Shallow Fusion with Large Language Models [62.76735265311028]
  We train a single multilingual language model (LM) for shallow fusion in multiple languages. Compared to a dense LM of similar computation during inference, GLaM reduces the WER of an English long-tail test set by 4.4% relative. In a multilingual shallow fusion task, GLaM improves 41 out of 50 languages with an average relative WER reduction of 3.85%, and a maximum reduction of 10%.
  arXiv  Detail & Related papers  (2023-02-17T14:46:38Z)
• Analyzing Bagging Methods for Language Models [0.5161531917413708]
  We perform an analysis of bagging language models and compare single language models to bagged ensembles that are roughly equivalent in terms of final model size. Our ensembling methods are at best roughly equivalent to single LM baselines.
  arXiv  Detail & Related papers  (2022-07-19T06:30:37Z)
• Finetuned Language Models Are Zero-Shot Learners [67.70352207685558]
  We show that instruction tuning boosts zero-shot performance on unseen tasks. We take a 137B parameter pretrained language model and instruction-tune it on over 60 NLP tasks verbalized via natural language instruction templates. We evaluate this instruction-tuned model, which we call FLAN, on unseen task types.
  arXiv  Detail & Related papers  (2021-09-03T17:55:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.