Pith. sign in

REVIEW 30 cited by

Repeat After Me: Transformers are Better than State Space Models at Copying

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 2402.01032 v2 pith:BIPRLVCH submitted 2024-02-01 cs.LG cs.AIcs.CL

Repeat After Me: Transformers are Better than State Space Models at Copying

classification cs.LG cs.AIcs.CL
keywords modelscopyinggssmsstatetransformerscontextspacetasks
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X LinkedIn Reddit HN
read the original abstract

Transformers are the dominant architecture for sequence modeling, but there is growing interest in models that use a fixed-size latent state that does not depend on the sequence length, which we refer to as "generalized state space models" (GSSMs). In this paper we show that while GSSMs are promising in terms of inference-time efficiency, they are limited compared to transformer models on tasks that require copying from the input context. We start with a theoretical analysis of the simple task of string copying and prove that a two layer transformer can copy strings of exponential length while GSSMs are fundamentally limited by their fixed-size latent state. Empirically, we find that transformers outperform GSSMs in terms of efficiency and generalization on synthetic tasks that require copying the context. Finally, we evaluate pretrained large language models and find that transformer models dramatically outperform state space models at copying and retrieving information from context. Taken together, these results suggest a fundamental gap between transformers and GSSMs on tasks of practical interest.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 30 Pith papers

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

  1. Can Transformers Really Do It All? On the Compatibility of Inductive Biases Across Tasks

    cs.LG 2026-07 conditional novelty 7.0

    Learned replacement non-linearities show transformers are rarely optimal for algorithmic tasks, with benefits that are task-specific, while language/code gains are smaller and more transferable.

  2. A Verifiable Search Is Not a Learnable Chain-of-Thought

    cs.LG 2026-06 unverdicted novelty 7.0

    Verifiable search procedures cannot be learned as forward chain-of-thought by language models; they instead learn memorization, verification, or require precomputed catalogs.

  3. Do Language Models Need Sleep? Offline Recurrence for Improved Online Inference

    cs.CL 2026-05 unverdicted novelty 7.0

    A sleep mechanism with N offline recurrent passes consolidates context into fast weights, improving performance on reasoning tasks where standard transformers fail.

  4. Positional LSH: Binary Block Matrix Approximation for Attention with Linear Biases

    cs.LG 2026-05 unverdicted novelty 7.0

    ALiBi bias is the expectation of positional LSH-induced block masks, yielding spectral and max-norm approximation bounds that reduce long-context biased attention to randomized short-context unbiased attention.

  5. The UNDO Flip-Flop: A Controlled Probe for Reversible Semantic State Management in State Space Model

    cs.LG 2026-04 unverdicted novelty 7.0

    Mamba-2 models fail to learn reversible state retrieval in the UNDO Flip-Flop task, defaulting to a toggle heuristic and achieving only 41% accuracy under adversarial conditions.

  6. The Bayesian Geometry of Transformer Attention

    cs.LG 2025-12 unverdicted novelty 7.0

    Small transformers reproduce known Bayesian posteriors with 10^{-3} to 10^{-4} bit accuracy in verifiable wind-tunnel tasks via residual belief states, FFN updates, and attention routing, while MLPs do not.

  7. Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models

    cs.LG 2024-02 unverdicted novelty 7.0

    Griffin hybrid model matches Llama-2 performance while trained on over 6 times fewer tokens and offers lower inference latency with higher throughput.

  8. DSSMs: State Space Models with Explicit Memory via Delay Differential Equations

    cs.LG 2026-07 conditional novelty 6.5

    Delay State Space Models augment diagonal SSMs with explicit delayed feedback, stable discrete parameterization, and FFT training, improving delayed-retrieval tasks and matching or beating S4D on most standard sequenc...

  9. DART: Decoded Attention over Recurrent States for Efficient Long-Context Sequence Modeling

    cs.LG 2026-08 conditional novelty 6.0

    DART augments Mamba-2 with attention over its own chunk state contributions, improving associative recall and retrieval with a smaller inference cache than token-level attention.

  10. Raven: High-Recall Sequence Modeling with Sparse Memory Routing

    cs.LG 2026-07 conditional novelty 6.0

    Raven is a linear-time sequence model whose sparse, input-dependent routing writes tokens into dedicated memory slots, preserving long-context recall and extrapolating 16x beyond training length.

  11. Kernelized Linear Attention: Breaking the Capacity Wall with Symmetric Cones

    cs.LG 2026-07 conditional novelty 6.0

    KATA uses rank-one PSD feature maps to pack exponentially many nearly orthogonal keys at fixed interference, reaching near-softmax MQAR at 16× length with about a quarter of softmax's KV-cache entries.

  12. The Capability Convergence Hypothesis: Capability from Access Structure, Not Scale

    cs.AI 2026-07 conditional novelty 6.0

    Capability, unlike representation, converges to access-complete hybrid architectures, and a witness task separates pure from hybrid families under fixed budgets.

  13. The Capability Convergence Hypothesis: Capability from Access Structure, Not Scale

    cs.AI 2026-07 conditional novelty 6.0

    Long-context capability converges to hybrid architectures pairing O(1) state with a scalable index; pure scaling of SSMs or Transformers alone hits information, horizon, or circuit walls.

  14. Remembering Distinct Items, Not Tokens: A Learnable Dirichlet-Process Cache Between State-Space Models and Attention

    cs.LG 2026-07 conditional novelty 6.0

    A DP-means allocate-on-novelty cache matches full-attention associative recall while storing only distinct items, and a minimal novelty gate recovers the rule end-to-end.

  15. A Hippocampus for Linear Attention: An Exact Memory for What the Recurrent State Forgets

    cs.AI 2026-07 unverdicted novelty 6.0

    HOLA pairs a compressive delta-rule recurrent state with a residual-selected exact KV cache and decoupled RMSNorm-gamma read, yielding lower perplexity than both standard linear attention and full-attention baselines ...

  16. Attention Amnesia in Hybrid LLMs: When CoT Fine-Tuning Breaks Long-Range Recall, and How to Fix It

    cs.CL 2026-06 conditional novelty 6.0

    CoT SFT disrupts long-range routing in hybrid models via changes to W_Q and W_K; QK-Restore restores pre-SFT projections to recover NIAH performance.

  17. Zamba2-VL Technical Report

    cs.CV 2026-05 unverdicted novelty 6.0

    Zamba2-VL is a family of 1.2B–7B hybrid Mamba2-transformer vision-language models that match leading transformer VLMs on image, reasoning, OCR, grounding and counting benchmarks while delivering roughly 10x lower time...

  18. OSDN: Improving Delta Rule with Provable Online Preconditioning in Linear Attention

    cs.LG 2026-05 unverdicted novelty 6.0

    OSDN adds online diagonal preconditioning to the Delta Rule, preserving chunkwise parallelism while proving super-geometric convergence and delivering 32-39% recall gains at 340M-1.3B scales.

  19. Echo: KV-Cache-Free Associative Recall with Spectral Koopman Operators

    cs.LG 2026-05 unverdicted novelty 6.0

    Spectral Koopman operators let SSMs achieve 100% accuracy on long-gap multi-query associative recall with fixed memory, where pure Mamba fails.

  20. The Recurrent Transformer: Greater Effective Depth and Efficient Decoding

    cs.LG 2026-04 unverdicted novelty 6.0

    Recurrent Transformers add per-layer recurrent memory via self-attention on own activations plus a tiling algorithm that reduces training memory traffic, yielding better C4 pretraining cross-entropy than parameter-mat...

  21. Towards Understanding What State Space Models Learn About Code

    cs.AI 2026-02 conditional novelty 6.0

    SSM code models capture code syntax and semantics better than Transformers before fine-tuning, forget short-range structure when fine-tuned on type inference, and an added high-frequency path or more kernels recovers ...

  22. Physics of Language Models: Part 4.1, Architecture Design and the Magic of Canon Layers

    cs.CL 2025-12 conditional novelty 6.0

    Canon layers—residual 1-d causal convolutions over adjacent tokens—boost synthetic reasoning depth 2-4x, lift NoPE to RoPE level, and bring GLA up to Mamba2/GDN, with qualitative real-world confirmation.

  23. Gated KalmaNet: A Fading Memory Layer Through Test-Time Ridge Regression

    cs.LG 2025-11 unverdicted novelty 6.0

    Gated KalmaNet uses exact Kalman gain computation with adaptive gating and Chebyshev iteration to improve SSM performance on long-context tasks over prior approximations like DeltaNet.

  24. Controllably Efficient Language Models

    cs.LG 2025-11 conditional novelty 6.0

    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.

  25. Kimi Linear: An Expressive, Efficient Attention Architecture

    cs.CL 2025-10 unverdicted novelty 6.0

    Kimi Linear hybridizes linear attention with a new KDA module to beat full attention on tasks while slashing KV cache by 75% and speeding decoding up to 6x.

  26. An Empirical Study of Mamba-based Language Models

    cs.LG 2024-06 accept novelty 6.0

    An 8B Mamba-2-Hybrid with 43% Mamba-2, 7% attention, and 50% MLP layers exceeds an 8B Transformer by 2.65 points on average across 12 tasks and matches it on 23 long-context tasks while enabling up to 8x faster inference.

  27. Context by Distinct Information: An Auditable Dirichlet-Process Working Memory for Long, Redundant Context Streams

    cs.LG 2026-07 conditional novelty 5.5

    A Dirichlet-process novelty cache plus recurrence matches full-attention coupling at about half the attended tokens on long redundant streams, with inspectable retained slots.

  28. Toeplitz MLP Mixers are Low Complexity, Information-Rich Sequence Models

    cs.LG 2026-04 unverdicted novelty 5.0

    Toeplitz MLP Mixers replace attention with masked Toeplitz multiplications for sub-quadratic complexity while retaining more sequence information and outperforming on copying and in-context tasks.

  29. TTT3R: 3D Reconstruction as Test-Time Training

    cs.CV 2025-09 unverdicted novelty 5.0

    TTT3R derives a closed-form learning rate from memory-observation alignment confidence to boost length generalization in RNN-based 3D reconstruction by 2x in global pose estimation.

  30. To model human linguistic prediction, make LLMs less superhuman

    cs.CL 2025-10 accept novelty 4.0

    LLMs' next-word predictions are too accurate to model human reading because their memory is superhuman, so cognitive modeling should build LLMs with human-like memory.