English
Related papers

Related papers: Understanding and Improving Length Generalization …

200 papers

Memory-augmented neural networks (MANNs) can perform algorithmic tasks such as sorting. However, they often fail to generalise to input sequence lengths not encountered during training. We introduce two approaches that constrain the state…

Machine Learning · Computer Science 2025-11-18 Patrick Ofner , Roman Kern

Broad textual understanding and in-context learning require language models that utilize full document contexts. Due to the implementation challenges associated with directly training long-context models, many methods have been proposed for…

Computation and Language · Computer Science 2024-09-24 Yi Lu , Jing Nathan Yan , Songlin Yang , Justin T. Chiu , Siyu Ren , Fei Yuan , Wenting Zhao , Zhiyong Wu , Alexander M. Rush

Recurrent Neural Networks (RNNs) offer fast inference on long sequences but are hard to optimize and slow to train. Deep state-space models (SSMs) have recently been shown to perform remarkably well on long sequence modeling tasks, and have…

Machine Learning · Computer Science 2023-03-14 Antonio Orvieto , Samuel L Smith , Albert Gu , Anushan Fernando , Caglar Gulcehre , Razvan Pascanu , Soham De

A prevailing narrative in LLM post-training holds that supervised finetuning (SFT) memorizes while reinforcement learning (RL) generalizes. We revisit this claim for reasoning SFT with long chain-of-thought (CoT) supervision and find that…

Artificial Intelligence · Computer Science 2026-04-09 Qihan Ren , Peng Wang , Ruikun Cai , Shuai Shao , Dadi Guo , Yuejin Xie , Yafu Li , Quanshi Zhang , Xia Hu , Jing Shao , Dongrui Liu

Transformer-based sequence-to-sequence architectures, while achieving state-of-the-art results on a large number of NLP tasks, can still suffer from overfitting during training. In practice, this is usually countered either by applying…

Computation and Language · Computer Science 2022-01-04 Dušan Variš , Ondřej Bojar

We extend the recent latent recurrent modeling to sequential input streams. By interleaving fast, recurrent latent updates with self-organizational ability between slow observation updates, our method facilitates the learning of stable…

Machine Learning · Computer Science 2026-04-23 Shota Takashiro , Masanori Koyama , Takeru Miyato , Yusuke Iwasawa , Yutaka Matsuo , Kohei Hayashi

In this study, we investigate the generalization of LSTM, ReLU and GRU models on counting tasks over long sequences. Previous theoretical work has established that RNNs with ReLU activation and LSTMs have the capacity for counting with…

Neural and Evolutionary Computing · Computer Science 2022-11-30 Nadine El-Naggar , Pranava Madhyastha , Tillman Weyde

In reinforcement learning, state representations are used to tractably deal with large problem spaces. State representations serve both to approximate the value function with few parameters, but also to generalize to newly encountered…

Machine Learning · Computer Science 2022-03-02 Charline Le Lan , Stephen Tu , Adam Oberman , Rishabh Agarwal , Marc G. Bellemare

Achieving effective test-time scaling requires models to engage in In-Context Exploration -- the intrinsic ability to generate, verify, and refine multiple reasoning hypotheses within a single continuous context. Grounded in State Coverage…

Computation and Language · Computer Science 2026-02-13 Futing Wang , Jianhao Yan , Yun Luo , Ganqu Cui , Zhi Wang , Xiaoye Qu , Yue Zhang , Yu Cheng , Tao Lin

Long-context modelling for large language models (LLMs) has been a key area of recent research because many real world use cases require reasoning over longer inputs such as documents. The focus of research into modelling long context has…

Computation and Language · Computer Science 2025-02-24 Wenhao Zhu , Pinzhen Chen , Hanxu Hu , Shujian Huang , Fei Yuan , Jiajun Chen , Alexandra Birch

Length Generalization is the essential capacity of autonomous agents to perform tasks in longer contexts than those encountered during training. To systematically study this feat, we test how well models can approximate the next token…

Re-initializing a neural network during training has been observed to improve generalization in recent works. Yet it is neither widely adopted in deep learning practice nor is it often used in state-of-the-art training protocols. This…

Machine Learning · Computer Science 2023-04-04 Sheheryar Zaidi , Tudor Berariu , Hyunjik Kim , Jörg Bornschein , Claudia Clopath , Yee Whye Teh , Razvan Pascanu

While state-of-the-art neural network models continue to achieve lower perplexity scores on language modeling benchmarks, it remains unknown whether optimizing for broad-coverage predictive performance leads to human-like syntactic…

Computation and Language · Computer Science 2020-05-26 Jennifer Hu , Jon Gauthier , Peng Qian , Ethan Wilcox , Roger P. Levy

A new breed of gated-linear recurrent neural networks has reached state-of-the-art performance on a range of sequence modeling problems. Such models naturally handle long sequences efficiently, as the cost of processing a new input is…

Machine Learning · Computer Science 2024-06-13 Maciej Pióro , Maciej Wołczyk , Razvan Pascanu , Johannes von Oswald , João Sacramento

Large language models (LLMs) often fail to scale their performance on long-context tasks performance in line with the context lengths they support. This gap is commonly attributed to retrieval failures -- the models' inability to identify…

Computation and Language · Computer Science 2025-10-08 Yufeng Du , Minyang Tian , Srikanth Ronanki , Subendhu Rongali , Sravan Bodapati , Aram Galstyan , Azton Wells , Roy Schwartz , Eliu A Huerta , Hao Peng

Despite their outstanding performance, large language models (LLMs) suffer notorious flaws related to their preference for simple, surface-level textual relations over full semantic complexity of the problem. This proposal investigates a…

Computation and Language · Computer Science 2022-06-20 Michal Štefánik

State-space models have gained popularity in sequence modelling due to their simple and efficient network structures. However, the absence of nonlinear activation along the temporal direction limits the model's capacity. In this paper, we…

Machine Learning · Computer Science 2023-11-02 Shida Wang , Beichen Xue

Memory retention challenges in deep neural architectures have ongoing limitations in the ability to process and recall extended contextual information. Token dependencies degrade as sequence length increases, leading to a decline in…

Computation and Language · Computer Science 2025-03-26 Frederick Dillon , Gregor Halvorsen , Simon Tattershall , Magnus Rowntree , Gareth Vanderpool

Recurrent Neural Networks (RNNs) continue to show outstanding performance in sequence modeling tasks. However, training RNNs on long sequences often face challenges like slow inference, vanishing gradients and difficulty in capturing long…

Artificial Intelligence · Computer Science 2018-02-06 Victor Campos , Brendan Jou , Xavier Giro-i-Nieto , Jordi Torres , Shih-Fu Chang

Simple recurrent neural networks (RNNs) and their more advanced cousins LSTMs etc. have been very successful in sequence modeling. Their theoretical understanding, however, is lacking and has not kept pace with the progress for feedforward…

Machine Learning · Computer Science 2021-06-02 Abhishek Panigrahi , Navin Goyal