We will typically read one or more papers per class session. You are responsible for reading all assigned papers. Additionally, you will submit a written review for each paper
Written paper reviews are a way to ensure you do the readings, and serve as the starting seed for class discussion. Anything you write in your review is fair game for me to bring up in class, so write as though all of your fellow students can see your reviews.
Reviews are due by noon the day before lecture. This gives me time to read what you all have written and to adjust the discussion accordingly. You are welcome to meet in small groups to discuss papers, but each student must submit his or her own review.
Each review has the following format. You will write four short paragraphs addressing the following points:
Stated goals and solution. What problem are the authors trying to solve? What are the bounds on this problem, i.e., what are they not trying to solve? What techniques or tools do the authors offer to solve the problem at hand? How do the authors know they have solved the problem? Do the authors test or validate their approach experimentally? Does the solution meet the stated goals, or does it fall short in some way? Avoid simply quoting the paper. Restating in your own words demonstrates your understanding.
Cool or significant ideas. What is new here? What are the main contributions of the paper? What did you find most interesting? Is this whole paper just a one-off clever trick or are there fundamental ideas here which could be reused in other contexts?
Fallacies and blind spots. Did the authors make any assumptions or disregard any issues that make their approach less appealing? Are there any theoretical problems, practical difficulties, implementation complexities, overlooked influences of evolving technology, and so on? Do you expect the technique to be more or less useful in the future? What kind of code or situation would defeat this approach, and are those programs or scenarios important in practice?
Note: critically reviewing a paper means more than just criticizing it. Think about strengths and weaknesses. Be realistic in your assessment. Don't expect a single paper to cover every possible issue or use case.
New ideas and connections to other work. How could the paper be extended? How could some of the flaws of the paper be corrected or avoided? Also, how does this paper relate to others we have read, or even any other research you are familiar with? Are there similarities between this approach and other work, or differences that highlight important facets of both?
Please take the time to edit your reviews. Unclear or unnecessarily long prose will be graded accordingly.
Please compose your review as plain ASCII text: no Microsoft Word, no PDF, just plain text. You will submit your reviews via the submit server :
Reviews will be graded on a four point scale:
Demonstrates exceptional insight about paper or related work. Noteworthy effort above and beyond just reading the assigned paper. Few if any 3-point grades will be awarded for each paper.
Clear and concise, demonstrating understanding of the key concepts of the paper. Ideas presented in your own words. Some evidence that the paper has been considered in the context of larger issues and themes of the course.
Shallow, minimally-sufficient, or needlessly wordy. Key concepts misunderstood or missing. Author’s words echoed back to me with little effort to reinterpret or paraphrase.
Late, incomplete, or never submitted at all.