proomt

Search

Search posts, papers, and topics

All posts

IBM ResearchSrinivasan Arunachalam, Arkopal Dutt, Hari Krovi, Rik Sengupta, Ryan Mandelbaum6 min readadvanced

A theoretical separation between quantum computers & LLMs

Read on ibm.com

Summary

The IBM research blog explains two new theoretical results that prove shallow constant‑depth quantum circuits can outperform decoder‑only transformers on a functional task (iterated index) and diffusion language models on a sampling task (parity‑sampling). The proofs give asymptotic separations but are not yet practical.

  • Proves a functional separation: shallow constant‑depth quantum circuits solve the iterated index problem while bounded‑resource decoder‑only transformers cannot.
  • Shows a sampling separation: constant‑depth quantum circuits efficiently sample parity‑constrained strings, a task diffusion language models with chain‑of‑thought cannot match.
  • Separations rely on transformer/DLM lower bounds and quantum upper bounds using a single classical AND gate.
  • Results are asymptotic and theoretical; they do not imply immediate advantage on today’s noisy quantum hardware.

Quantum algorithm researchers and AI engineers should know that even very shallow quantum circuits can provably exceed the capabilities of current LLM architectures on specific tasks, informing future benchmark and hybrid system design.

6/10
transformerstheoryquantum computinglarge language modelscomplexity

Related reading

  1. From error mitigation to fault-tolerant quantum computing

    IBM’s blog outlines a practical continuum from quantum error mitigation to fault‑tolerant error correction, highlighting hybrid spacetime codes that achieve ~10× effective error‑rate reduction (64 logical qubits on 76 physical qubits) and a 63× sampling‑overhead cut versus pure mitigation. It describes how post‑selected QEC combined with probabilistic error cancellation, dynamic circuits, and a n…

    IBM Researchibm.com10 min

  3. Pruning LLMs Like a Physicist: Block Removal as an Ising Optimization Problem

    The authors cast transformer block removal as a constrained binary optimization problem equivalent to an Ising glass, using a Hessian‑derived energy as a proxy for downstream quality. Solving the resulting QUBO with classical or quantum‑inspired solvers yields up to 23 MMLU points improvement over prior block‑removal baselines at 50 % depth compression.

    Hugging Facehuggingface.co8 min

  4. How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus

    This paper investigates the "lossless" claim of Orthrus, a hybrid architecture for accelerating LLM inference. It finds that under BF16 precision, Orthrus diverges from the exact autoregressive output trajectory in over 50% of cases, though FP32 maintains exact matching. Despite BF16 divergence, downstream task performance was not systematically degraded.

    Hugging Face Daily Papersarxiv.org1 minpaper

  5. Fathom: Per-Query Read Depth for Sparse Decoding over Offloaded KV Caches

    Fathom introduces a per-query read depth mechanism for sparse decoding over offloaded KV caches, allowing each query to adaptively decide how many bits of each key channel to read. This method significantly speeds up decoding for large language models with long contexts by reducing host memory traffic, achieving 1.67x faster GPU decoding on Qwen3-8B at one million tokens.

    Hugging Face Daily Papersarxiv.org1 minpaper

  6. Introducing System One Models and Jev

    TypeSafe AI announced its first “System One” model, Jev, a non‑text‑generating LLM that outputs type‑safe structured decisions with calibrated probabilities. It claims 40‑200× lower latency (70‑500 ms) and 100‑500× lower cost versus frontier LLMs, no hallucinations, and parallel sampling. The post includes a side‑by‑side demo, a custom “workflow” benchmark comparing Jev to GPT‑5.6/6 and other mod…

    Hacker News front pagetypesafe.ai9 minHN1824480lobste.rs26