SCIT: Testing Causal Cache Carriers in Latent Chain-of-Thought Models

• URL: http://arxiv.org/abs/2608.27265v1
• Date: Thu, 27 Aug 2026 15:47:48 GMT
• Title: SCIT: Testing Causal Cache Carriers in Latent Chain-of-Thought Models
• Abstract summary: We introduce SCIT, a causal protocol that constructs exact source-recipient counterfactuals.<n>On CODI-GPT2 and a Sim-CoT-style GPT-2 reproduction, counterfactual arithmetic transfers primarily through value-cache suffix trajectories.
• Score: 7.892321765328599
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Latent chain-of-thought models move intermediate reasoning from emitted text into continuous states, improving compactness but hiding the causal object. We introduce SCIT, the Suffix Cache Interchange Test, a causal protocol that constructs exact source-recipient counterfactuals, patches declared cache segments, and identifies which transformer object carries the counterfactual computation. SCIT combines sufficiency tests with K/V component splits, hidden-state controls, semantic source controls, decoded validation, and matched corruption. On CODI-GPT2 and a Sim-CoT-style GPT-2 reproduction, counterfactual arithmetic transfers primarily through value-cache suffix trajectories rather than hidden states, keys, reusable answer slots, or single-token triggers. Complete sufficiency-and-necessity evidence for the late-value-suffix mechanism holds for the main CODI-GPT2 checkpoint; the Sim-CoT-style checkpoint shows the same sufficiency and decoded-control pattern but insufficient matched-corruption evidence for a necessity call. Beyond these local arithmetic cells, SCIT reveals carrier-regime shifts: arithmetic-like GPT-2/1B cells preserve latent-tail value/KV transfer, whereas competent 8B and repaired non-arithmetic cells route through prompt-prefix or full-cache K/V; boundary cells receive no mechanism call. SCIT therefore contributes a cache-level diagnostic, a checkpoint-specific GPT-2 arithmetic mechanism, and a competence-gated carrier map rather than a universal latent-tail claim.

Related papers

• CateKV: On Sequential Consistency for Long-Context LLM Inference Acceleration [67.17709877327232]
  Cate KV is a hybrid KV cache method that retains only critical token information for consistent heads.<n>Cate KV reduces memory usage by up to $2.72times$ and accelerates decoding by $2.18times$ in single-sample inputs.
  arXiv  Detail & Related papers  (2026-08-31T06:02:37Z)
• Quantization-Triggered Backdoors in Language Models: Cross-Quantizer Transferability and the Validation--Deployment Gap [0.0]
  Post-training quantization is often treated as a semantically neutral optimization for edge deployment of Large Language Models.<n>We show that backdoored translation models move from zero measured friend--foe corruption at repaired FP16 to up to 85.02% inversion after quantization.<n>We formalize this gap through Quantization Behavioral Equivalence Classes (QBECs) and prove that QBEC membership does not imply behavioral equivalence.
  arXiv  Detail & Related papers  (2026-08-27T08:34:10Z)
• Back from the Future: Key-Value Cache Management by Counter-Causal Surprise [67.1056509495879]
  Key-value (KV) cache management through compression and eviction strategies has emerged as an important research direction in recent years.<n>We propose a simple yet effective KV eviction scheme motivated by the insight that past tokens which can be well-predicted from more recent tokens are redundant.<n>We evaluate our strategy on various open-source LLMs and benchmark datasets showing competitive or improved performance over other state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-07-30T02:42:51Z)
• Remembering Distinct Items, Not Tokens: A Learnable Dirichlet-Process Cache Between State-Space Models and Attention [0.48342038441006796]
  We study a sparse cache that allocates a slot only when an input is novel, so its size tracks the number of distinct items rather than the number of tokens.<n>We develop it in two forms, a static cache with a fixed concentration and a surprise-adaptive variant whose concentration follows the recent novelty rate.
  arXiv  Detail & Related papers  (2026-07-10T18:28:21Z)
• Closing the Null Space: Guidance-Aware Quantization for Classifier-Free Diffusion [0.0]
  Two-pass CFG execution pattern imposes a latency overhead that parameter-count and bit-operation metrics conceal entirely.<n>Guidance-Aware Mixed Precision (GAMP) is proposed, which calibrates directly on the guided prediction.
  arXiv  Detail & Related papers  (2026-07-09T08:42:07Z)
• Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG [17.045113249152234]
  We introduce the first end-to-end side-channel attack targeting non- KV cache fusion.<n>SpliceLeak executes a systematic two-phase privacy breach.<n>SpliceDefense effectively flattens the side-channel signal (Delta TTFT 0) with negligible throughput overhead.
  arXiv  Detail & Related papers  (2026-06-20T02:28:28Z)
• SIM-CoT: Supervised Implicit Chain-of-Thought [108.30049193668083]
  Implicit Chain-of-Thought (CoT) methods offer a token-efficient alternative to explicit CoT reasoning in Large Language Models.<n>We identify a core latent instability issue when scaling the computational budget of implicit CoT.<n>We propose SIM-CoT, a plug-and-play training module that introduces step-level supervision to stabilize and enrich the latent reasoning space.
  arXiv  Detail & Related papers  (2025-09-24T17:01:32Z)
• Efficient Blockchain-based Steganography via Backcalculating Generative Adversarial Network [105.47203971578871]
  We propose a generic blockchain-based steganography framework (GBSF)<n>The sender generates the required fields such as amount and fees, where the additional covert data is embedded to enhance the channel capacity.<n>Based on GBSF, we design a reversible generative adversarial network (R-GAN)<n>We propose R-GAN with Counter-intuitive data preprocessing and Custom activation functions, namely CCR-GAN.
  arXiv  Detail & Related papers  (2025-06-19T04:43:41Z)
• Semi-DETR: Semi-Supervised Object Detection with Detection Transformers [105.45018934087076]
  We analyze the DETR-based framework on semi-supervised object detection (SSOD) We present Semi-DETR, the first transformer-based end-to-end semi-supervised object detector. Our method outperforms all state-of-the-art methods by clear margins.
  arXiv  Detail & Related papers  (2023-07-16T16:32:14Z)
• Quantum Proofs of Deletion for Learning with Errors [91.3755431537592]
  We construct the first fully homomorphic encryption scheme with certified deletion. Our main technical ingredient is an interactive protocol by which a quantum prover can convince a classical verifier that a sample from the Learning with Errors distribution in the form of a quantum state was deleted.
  arXiv  Detail & Related papers  (2022-03-03T10:07:32Z)

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.