Leonardo Kuffo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Leonardo Kuffo has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Peter Boncz
CWI, Amsterdam, Netherlands
4 shared publications - Azim Afroozeh
CWI, Amsterdam, Netherlands
2 shared publications - Albert Angel
Google, Zurich, Switzerland
1 shared publication - Bogdan Nicuşor Bindea
Department of Computer Science, Technical University of Cluj-Napoca, 400027 Cluj-Napoca, Romania
1 shared publication - Carmen Vaca
Escuela Superior Politécnica del Litoral (ESPOL)
1 shared publication - Elena Krippner
CWI, Amsterdam, Netherlands
1 shared publication
Latest Publications
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 3907-3912
Semantic Recall for Vector Search
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 3907-3912
Open AccessSemantic Recall for Vector Search
- Open Access
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Bang for the Buck: Vector Search on Cloud CPUs
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessBang for the Buck: Vector Search on Cloud CPUs
- Open Access
- 17 June 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 3 (3) , 1-26
PDX: A Data Layout for Vector Similarity Search
We propose Partition Dimensions Across (PDX), a data layout for vectors (e.g., embeddings) that, similar to PAX [6], stores multiple vectors in one block, using a vertical layout for the dimensions (Figure 1). PDX accelerates exact and approximate similarity search thanks to its dimension-by-dimension search strategy that operates on multiple-vectors-at-a-time in tight loops. It beats SIMD-optimized distance kernels on standard horizontal vector storage (avg 40% faster), only relying on scalar code that gets auto-vectorized. We combined the PDX layout with recent dimension-pruning algorithms ADSampling [19] and BSA [52] that accelerate approximate vector search. We found that these algorithms on the horizontal vector layout can lose to SIMD-optimized linear scans, even if they are SIMD-optimized. However, when used on PDX, their benefit is restored to 2-7x. We find that search on PDX is especially fast if a limited number of dimensions has to be scanned fully, which is what the dimension-pruning approaches do. We finally introduce PDX-BOND, an even more flexible dimension-pruning strategy, with good performance on exact search and reasonable performance on approximate search. Unlike previous pruning algorithms, it can work on vector data ''as-is'' without preprocessing; making it attractive for vector databases with frequent updates.
- 17 June 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 3 (3) , 1-26
Open AccessPDX: A Data Layout for Vector Similarity Search
We propose Partition Dimensions Across (PDX), a data layout for vectors (e.g., embeddings) that, similar to PAX [6], stores multiple vectors in one block, using a vertical layout for the dimensions (Figure 1). PDX accelerates exact and approximate similarity search thanks to its dimension-by-dimension search strategy that operates on multiple-vectors-at-a-time in tight loops. It beats SIMD-optimized distance kernels on standard horizontal vector storage (avg 40% faster), only relying on scalar code that gets auto-vectorized. We combined the PDX layout with recent dimension-pruning algorithms ADSampling [19] and BSA [52] that accelerate approximate vector search. We found that these algorithms on the horizontal vector layout can lose to SIMD-optimized linear scans, even if they are SIMD-optimized. However, when used on PDX, their benefit is restored to 2-7x. We find that search on PDX is especially fast if a limited number of dimensions has to be scanned fully, which is what the dimension-pruning approaches do. We finally introduce PDX-BOND, an even more flexible dimension-pruning strategy, with good performance on exact search and reasonable performance on approximate search. Unlike previous pruning algorithms, it can work on vector data ''as-is'' without preprocessing; making it attractive for vector databases with frequent updates.
- Open Access
- 21 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 51-54
Reproducibility Report for ACM SIGMOD 2024 Paper: 'ALP: Adaptive Lossless Floating-Point Compression'
- 21 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 51-54
Open AccessReproducibility Report for ACM SIGMOD 2024 Paper: 'ALP: Adaptive Lossless Floating-Point Compression'
- Top Cited PapersOpen Access
- 8 December 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 1 (4) , 1-26
ALP: Adaptive Lossless floating-Point Compression
IEEE 754 doubles do not exactly represent most real values, introducing rounding errors in computations and [de]serialization to text. These rounding errors inhibit the use of existing lightweight compression schemes such as Delta and Frame Of Reference (FOR), but recently new schemes were proposed: Gorilla, Chimp128, PseudoDecimals (PDE), Elf and Patas. However, their compression ratios are not better than those of general-purpose compressors such as Zstd; while [de]compression is much slower than Delta and FOR. We propose and evaluate ALP, that significantly improves these previous schemes in both speed and compression ratio (Figure 1). We created ALP after carefully studying the datasets used to evaluate the previous schemes. To obtain speed, ALP is designed to fit vectorized execution. This turned out to be key for also improving the compression ratio, as we found in-vector commonalities to create compression opportunities. ALP is an adaptive scheme that uses a strongly enhanced version of PseudoDecimals [31] to losslessly encode doubles as integers if they originated as decimals, and otherwise uses vectorized compression of the doubles' front bits. Its high speeds stem from our implementation in scalar code that auto-vectorizes, using building blocks provided by our FastLanes library [6], and an efficient two-stage compression algorithm that first samples row-groups and then vectors.
- 8 December 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 1 (4) , 1-26
Top Cited PapersOpen AccessALP: Adaptive Lossless floating-Point Compression
IEEE 754 doubles do not exactly represent most real values, introducing rounding errors in computations and [de]serialization to text. These rounding errors inhibit the use of existing lightweight compression schemes such as Delta and Frame Of Reference (FOR), but recently new schemes were proposed: Gorilla, Chimp128, PseudoDecimals (PDE), Elf and Patas. However, their compression ratios are not better than those of general-purpose compressors such as Zstd; while [de]compression is much slower than Delta and FOR. We propose and evaluate ALP, that significantly improves these previous schemes in both speed and compression ratio (Figure 1). We created ALP after carefully studying the datasets used to evaluate the previous schemes. To obtain speed, ALP is designed to fit vectorized execution. This turned out to be key for also improving the compression ratio, as we found in-vector commonalities to create compression opportunities. ALP is an adaptive scheme that uses a strongly enhanced version of PseudoDecimals [31] to losslessly encode doubles as integers if they originated as decimals, and otherwise uses vectorized compression of the doubles' front bits. Its high speeds stem from our implementation in scalar code that auto-vectorizes, using building blocks provided by our FastLanes library [6], and an efficient two-stage compression algorithm that first samples row-groups and then vectors.
- 21 July 2021
- Vol. 1
Automated CX Detection through Social Networks - Scripts & Dataset
- 1 April 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 148-153
Back to #6D: Predicting Venezuelan states political election results through Twitter
The large adoption of Twitter during electioneering has created an unprecedented opportunity to capture the citizen's behaviour nationwide. The real-time access to information published by citizens has motivated researchers to design methods in order to enrich traditional political polling with insights from this rich source of data. However, less work has been done to capture the political scenario in Latin American countries, given that some methods rely on the use of English words, the reproducibility of such studies in Spanish speaking countries is a challenging task. Therefore, we propose a framework in which we apply social network analysis techniques and unsupervised machine learning to infer the political alignment at state level during Venezuelan Parliamentary election, which were performed on December 6, 2015. This electoral process took place in the middle of an acute political polarization in the country, the masses were organized around two political coalitions with opposite ideology: Government and opposition. In order to discover automatically the corresponding state political preferences, we analyze 60K tweets posted within the Venezuelan geographic boundaries during one week before the election day. Applying our framework, we are able to infer a given state political alignment starting from the quantified differences in communication patterns and linguistic profiles of the state aggregated tweets. We demonstrate that the online political atmosphere reflects the offline tendency at state scale given that we are able to predict the election tendency in Venezuela states with an accuracy of 87.5% with respect to official election results publicly available.
- 1 April 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 148-153
Back to #6D: Predicting Venezuelan states political election results through Twitter
The large adoption of Twitter during electioneering has created an unprecedented opportunity to capture the citizen's behaviour nationwide. The real-time access to information published by citizens has motivated researchers to design methods in order to enrich traditional political polling with insights from this rich source of data. However, less work has been done to capture the political scenario in Latin American countries, given that some methods rely on the use of English words, the reproducibility of such studies in Spanish speaking countries is a challenging task. Therefore, we propose a framework in which we apply social network analysis techniques and unsupervised machine learning to infer the political alignment at state level during Venezuelan Parliamentary election, which were performed on December 6, 2015. This electoral process took place in the middle of an acute political polarization in the country, the masses were organized around two political coalitions with opposite ideology: Government and opposition. In order to discover automatically the corresponding state political preferences, we analyze 60K tweets posted within the Venezuelan geographic boundaries during one week before the election day. Applying our framework, we are able to infer a given state political alignment starting from the quantified differences in communication patterns and linguistic profiles of the state aggregated tweets. We demonstrate that the online political atmosphere reflects the offline tendency at state scale given that we are able to predict the election tendency in Venezuela states with an accuracy of 87.5% with respect to official election results publicly available.