Thierry Tambe
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Thierry Tambe has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Gu-Yeon Wei
Harvard University, Cambridge, MA, USA
4 shared publications - Aarti Gupta
Princeton University, Princeton, USA
2 shared publications - Akash Gaonkar
Princeton University, USA
2 shared publications - Andrew Cheung
University of Washington, Seattle, USA
2 shared publications - Bo-Yuan Huang
Department of Electrical Engineering, Princeton University, Princeton, NJ, USA
2 shared publications - David Brooks
Harvard University, Cambridge, MA, USA
2 shared publications
Latest Publications
- Open Access
- 16 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Solid-State Circuits Society
- Vol. PP, 1
A 18.1TFLOPs/W Transformer Accelerator with Fine-Grained Per-Query Latency and Power Management in 12-nm FinFET
The transformer neural architecture has catalyzed a series of AI breakthroughs in the field of natural language processing, fueling the growth of AI conversational interfaces. The quality of service for such AI-based chatbots is typically defined by the round-trip latency following a user’s prompt and the energy consumption of the computing hardware. We present a sparse transformer processor, named Atropos, that tailors its latency and energy expenditures according to the complexity of the input prompt it processes. The 12nm processor introduces three key innovations: (1) a specialized datapath for entropy-based early exit assessment that reduces BERT inference latency by up to 6.13×, (2) a mixed-precision FP4/FP8 MAC unit with per-vector exponent biases that doubles throughput while maintaining accuracy, and (3) a prompt-level power management scheme that dynamically scales voltage and frequency based on input prompt complexity. These innovations together enable the STP to achieve a peak efficiency of 18.1 TFLOPs/W and 65mJ per inference, representing a 7.14× energy improvement over conventional BERT inference.
- 16 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Solid-State Circuits Society
- Vol. PP, 1
Open AccessA 18.1TFLOPs/W Transformer Accelerator with Fine-Grained Per-Query Latency and Power Management in 12-nm FinFET
The transformer neural architecture has catalyzed a series of AI breakthroughs in the field of natural language processing, fueling the growth of AI conversational interfaces. The quality of service for such AI-based chatbots is typically defined by the round-trip latency following a user’s prompt and the energy consumption of the computing hardware. We present a sparse transformer processor, named Atropos, that tailors its latency and energy expenditures according to the complexity of the input prompt it processes. The 12nm processor introduces three key innovations: (1) a specialized datapath for entropy-based early exit assessment that reduces BERT inference latency by up to 6.13×, (2) a mixed-precision FP4/FP8 MAC unit with per-vector exponent biases that doubles throughput while maintaining accuracy, and (3) a prompt-level power management scheme that dynamically scales voltage and frequency based on input prompt complexity. These innovations together enable the STP to achieve a peak efficiency of 18.1 TFLOPs/W and 65mJ per inference, representing a 7.14× energy improvement over conventional BERT inference.
- Open Access
- 13 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 27-36
Towards Memory Specialization: A Case for Long-Term and Short-Term RAM
- 13 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 27-36
Open AccessTowards Memory Specialization: A Case for Long-Term and Short-Term RAM
- Open Access
- 5 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
JointNF: Enhancing DNN Performance through Adaptive N:M Pruning across both Weight and Activation
Balancing accuracy and hardware efficiency remains a challenge with traditional pruning methods. N:M sparsity is a recent approach offering a compromise, allowing up to N non-zero weights in a group of M consecutive weights. However, N:M pruning enforces a uniform sparsity level of N/M across all layers, which does not align well sparse nature of deep neural networks (DNNs). To achieve a more flexible sparsity pattern and a higher overall sparsity level, we present JointNF, a novel joint N:M and structured pruning algorithm to enable fine-grained structured pruning with adaptive sparsity levels across the DNN layers. Moreover, we show for the first time that N:M pruning can also be applied over the input activation for further performance enhancement.
- 5 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessJointNF: Enhancing DNN Performance through Adaptive N:M Pruning across both Weight and Activation
Balancing accuracy and hardware efficiency remains a challenge with traditional pruning methods. N:M sparsity is a recent approach offering a compromise, allowing up to N non-zero weights in a group of M consecutive weights. However, N:M pruning enforces a uniform sparsity level of N/M across all layers, which does not align well sparse nature of deep neural networks (DNNs). To achieve a more flexible sparsity pattern and a higher overall sparsity level, we present JointNF, a novel joint N:M and structured pruning algorithm to enable fine-grained structured pruning with adaptive sparsity levels across the DNN layers. Moreover, we show for the first time that N:M pruning can also be applied over the input activation for further performance enhancement.
- Open Access
- 14 February 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 29 (2) , 1-25
Application-level Validation of Accelerator Designs Using a Formal Software/Hardware Interface
Ideally, accelerator development should be as easy as software development. Several recent design languages/tools are working toward this goal, but actually testing early designs on real applications end-to-end remains prohibitively difficult due to the costs of building specialized compiler and simulator support. We propose a new first-in-class, mostly automated methodology termed “3LA” to enable end-to-end testing of prototype accelerator designs on unmodified source applications. A key contribution of 3LA is the use of a formal software/hardware interface that specifies an accelerator’s operations and their semantics. Specifically, we leverage the Instruction-level Abstraction (ILA) formal specification for accelerators that has been successfully used thus far for accelerator implementation verification. We show how the ILA for accelerators serves as a software/hardware interface, similar to the Instruction Set Architecture for processors, that can be used for automated development of compilers and instruction-level simulators. Another key contribution of this work is to show how ILA-based accelerator semantics enables extending recent work on equality saturation to auto-generate basic compiler support for prototype accelerators in a technique we term “flexible matching.” By combining flexible matching with simulators auto-generated from ILA specifications, our approach enables end-to-end evaluation with modest engineering effort. We detail several case studies of 3LA, which uncovered an unknown flaw in a recently published accelerator and facilitated its fix.
- 14 February 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 29 (2) , 1-25
Open AccessApplication-level Validation of Accelerator Designs Using a Formal Software/Hardware Interface
Ideally, accelerator development should be as easy as software development. Several recent design languages/tools are working toward this goal, but actually testing early designs on real applications end-to-end remains prohibitively difficult due to the costs of building specialized compiler and simulator support. We propose a new first-in-class, mostly automated methodology termed “3LA” to enable end-to-end testing of prototype accelerator designs on unmodified source applications. A key contribution of 3LA is the use of a formal software/hardware interface that specifies an accelerator’s operations and their semantics. Specifically, we leverage the Instruction-level Abstraction (ILA) formal specification for accelerators that has been successfully used thus far for accelerator implementation verification. We show how the ILA for accelerators serves as a software/hardware interface, similar to the Instruction Set Architecture for processors, that can be used for automated development of compilers and instruction-level simulators. Another key contribution of this work is to show how ILA-based accelerator semantics enables extending recent work on equality saturation to auto-generate basic compiler support for prototype accelerators in a technique we term “flexible matching.” By combining flexible matching with simulators auto-generated from ILA specifications, our approach enables end-to-end evaluation with modest engineering effort. We detail several case studies of 3LA, which uncovered an unknown flaw in a recently published accelerator and facilitated its fix.
- Open Access
Application-Level Validation of Accelerator Designs Using a Formal Software/Hardware Interface
Ideally, accelerator development should be as easy as software development. Several recent design languages/tools are working toward this goal, but actually testing early designs on real applications end-to-end remains prohibitively difficult due to the costs of building specialized compiler and simulator support. We propose a new first-in-class, mostly automated methodology termed "3LA" to enable end-to-end testing of prototype accelerator designs on unmodified source applications. A key contribution of 3LA is the use of a formal software/hardware interface that specifies an accelerator's operations and their semantics. Specifically, we leverage the Instruction-Level Abstraction (ILA) formal specification for accelerators that has been successfully used thus far for accelerator implementation verification. We show how the ILA for accelerators serves as a software/hardware interface, similar to the Instruction Set Architecture (ISA) for processors, that can be used for automated development of compilers and instruction-level simulators. Another key contribution of this work is to show how ILA-based accelerator semantics enables extending recent work on equality saturation to auto-generate basic compiler support for prototype accelerators in a technique we term "flexible matching." By combining flexible matching with simulators auto-generated from ILA specifications, our approach enables end-to-end evaluation with modest engineering effort. We detail several case studies of 3LA, which uncovered an unknown flaw in a recently published accelerator and facilitated its fix.
Open AccessApplication-Level Validation of Accelerator Designs Using a Formal Software/Hardware Interface
Ideally, accelerator development should be as easy as software development. Several recent design languages/tools are working toward this goal, but actually testing early designs on real applications end-to-end remains prohibitively difficult due to the costs of building specialized compiler and simulator support. We propose a new first-in-class, mostly automated methodology termed "3LA" to enable end-to-end testing of prototype accelerator designs on unmodified source applications. A key contribution of 3LA is the use of a formal software/hardware interface that specifies an accelerator's operations and their semantics. Specifically, we leverage the Instruction-Level Abstraction (ILA) formal specification for accelerators that has been successfully used thus far for accelerator implementation verification. We show how the ILA for accelerators serves as a software/hardware interface, similar to the Instruction Set Architecture (ISA) for processors, that can be used for automated development of compilers and instruction-level simulators. Another key contribution of this work is to show how ILA-based accelerator semantics enables extending recent work on equality saturation to auto-generate basic compiler support for prototype accelerators in a technique we term "flexible matching." By combining flexible matching with simulators auto-generated from ILA specifications, our approach enables end-to-end evaluation with modest engineering effort. We detail several case studies of 3LA, which uncovered an unknown flaw in a recently published accelerator and facilitated its fix.
- 8 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 58 (2) , 569-581
A 16-nm SoC for Noise-Robust Speech and NLP Edge AI Inference With Bayesian Sound Source Separation and Attention-Based DNNs
The proliferation of personal artificial intelligence (AI) -assistant technologies with speech-based conversational AI interfaces is driving the exponential growth in the consumer Internet of Things (IoT) market. As these technologies are being applied to keyword spotting (KWS), automatic speech recognition (ASR), natural language processing (NLP), and text-to-speech (TTS) applications, it is of paramount importance that they provide uncompromising performance for context learning in long sequences, which is a key benefit of the attention mechanism, and that they work seamlessly in polyphonic environments. In this work, we present a 25-mm$^2$ system-on-chip (SoC) in 16-nm FinFET technology, codenamed SM6, which executes end-to-end speech-enhancing attention-based ASR and NLP workloads. The SoC includes: 1) FlexASR, a highly reconfigurable NLP inference processor optimized for whole-model acceleration of bidirectional attention-based sequence-to-sequence (seq2seq) deep neural networks (DNNs); 2) a Markov random field source separation engine (MSSE), a probabilistic graphical model accelerator for unsupervised inference via Gibbs sampling, used for sound source separation; 3) a dual-core Arm Cortex A53 CPU cluster, which provides on-demand single Instruction/multiple data (SIMD) fast fourier transform (FFT) processing and performs various application logic (e.g., expectation–maximization (EM) algorithm and 8-bit floating-point (FP8) quantization); and 4) an always-on M0 subsystem for audio detection and power management. Measurement results demonstrate the efficiency ranges of 2.6–7.8 TFLOPs/W and 4.33–17.6 Gsamples/s/W for FlexASR and MSSE, respectively; MSSE denoising performance allowing 6$\times$ smaller ASR model to be stored on-chip with negligible accuracy loss; and 2.24-mJ energy consumption while achieving real-time throughput, end-to-end, and per-frame ASR latencies of 18 ms.- 8 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Solid-State Circuits
- Vol. 58 (2) , 569-581
A 16-nm SoC for Noise-Robust Speech and NLP Edge AI Inference With Bayesian Sound Source Separation and Attention-Based DNNs
The proliferation of personal artificial intelligence (AI) -assistant technologies with speech-based conversational AI interfaces is driving the exponential growth in the consumer Internet of Things (IoT) market. As these technologies are being applied to keyword spotting (KWS), automatic speech recognition (ASR), natural language processing (NLP), and text-to-speech (TTS) applications, it is of paramount importance that they provide uncompromising performance for context learning in long sequences, which is a key benefit of the attention mechanism, and that they work seamlessly in polyphonic environments. In this work, we present a 25-mm$^2$ system-on-chip (SoC) in 16-nm FinFET technology, codenamed SM6, which executes end-to-end speech-enhancing attention-based ASR and NLP workloads. The SoC includes: 1) FlexASR, a highly reconfigurable NLP inference processor optimized for whole-model acceleration of bidirectional attention-based sequence-to-sequence (seq2seq) deep neural networks (DNNs); 2) a Markov random field source separation engine (MSSE), a probabilistic graphical model accelerator for unsupervised inference via Gibbs sampling, used for sound source separation; 3) a dual-core Arm Cortex A53 CPU cluster, which provides on-demand single Instruction/multiple data (SIMD) fast fourier transform (FFT) processing and performs various application logic (e.g., expectation–maximization (EM) algorithm and 8-bit floating-point (FP8) quantization); and 4) an always-on M0 subsystem for audio detection and power management. Measurement results demonstrate the efficiency ranges of 2.6–7.8 TFLOPs/W and 4.33–17.6 Gsamples/s/W for FlexASR and MSSE, respectively; MSSE denoising performance allowing 6$\times$ smaller ASR model to be stored on-chip with negligible accuracy loss; and 2.24-mJ energy consumption while achieving real-time throughput, end-to-end, and per-frame ASR latencies of 18 ms.